Social organisation and population genetics of the Threatened great desert skink, Liopholis kintorei

Siobhan Dennison BSc (Hons I)

Department of Biological Sciences Faculty of Science and Engineering Macquarie University

Thesis presented for the degree of Doctor of Philosophy, March 2015

TABLE OF CONTENTS

ABSTRACT...... i DECLARATION...... iii PERSONAL ACKNOWLEDGEMENTS...... iv PRESENTATIONS OF RESEARCH FROM THIS THESIS...... vii LIST OF FIGURES...... viii LIST OF TABLES...... x

CHAPTER 1: General introduction...... 1

Social organisation...... 2 Social organisation in reptiles...... 3 The Egernia group as a model system...... 5 Thesis overview...... 9 Literature cited...... 11

CHAPTER 2: Genetic divergence among regions containing the Vulnerable great desert skink (Liopholis kintorei) in the Australian arid zone...... 19

Abstract...... 20 Introduction...... 21 Methods...... 23 Results...... 28 Discussion...... 34 Acknowledgements...... 38 Literature cited...... 39 Supplementary material 2.1 – microsatellite development...... 46 Supplementary material 2.2 – genetic divergence plots...... 50 Supplementary material 2.3 – climate data...... 52

CHAPTER 3: Sex-biased dispersal and fine-scale movement in a group- living lizard, the great desert skink, Liopholis kintorei...... 55

Abstract...... 56 Introduction...... 56 Methods...... 63 Results...... 67 Discussion...... 72 Acknowledgements...... 77 Literature cited...... 78 Supplementary material 3.1 – additional analyses of movement...... 86

!

CHAPTER 4: Variable social organisation in a group-living lizard, Liopholis kintorei...... 87

Abstract...... 88 Introduction...... 89 Methods...... 93 Results...... 99 Discussion...... 109 Acknowledgements...... 118 Literature cited...... 118 Supplementary material 4.1 – growth curve...... 126 Supplementary material 4.2 – additional analyses of relatedness...... 129

CHAPTER 5: Do bigger heads make burly males? Sexual selection and the evolution of sexual head size dimorphism in lizards...... 137

Abstract...... 138 Introduction...... 138 Methods...... 143 Results...... 146 Discussion...... 154 Acknowledgements...... 159 Literature cited...... 159 Supplementary material 5.1 – correlation coefficients...... 173

CHAPTER 6: General Discussion...... 175

Overview...... 176 Evolutionary implications...... 179 Conservation implications...... 182 Suggestions for future research...... 184 Conclusion...... 188 Literature cited...... 189

APPENDIX: Additional information...... 195

! ! ! Abstract

The great desert skink, Liopholis kintorei, is a listed Threatened species with a patchy distribution across the arid zone of central and western Australia. It is a member of a group of Australian lizards, the Egernia group, recently adopted as model species for the early evolution and maintenance of vertebrate sociality. Most

Egernia group species that live in stable, kin-based groups utilise pre-existing shelter sites such as rock crevices, but L. kintorei is unique in that individuals cooperatively construct and maintain extensive burrow systems that house kin.

I researched aspects of the behavioural ecology and social structure of L. kintorei, using field-based methods and molecular techniques. I captured individuals from a study population at the Australian Wildlife Conservancy’s Newhaven

Sanctuary in central Australia, at which a total of seven months of fieldwork was carried out over three years. The results of my research are discussed in the context of the evolution and maintenance of kin-based social living, with comments on the implications for the species’ conservation in central Australia.

Social organisation was investigated by trapping lizards at their burrows over the three years and through molecular analyses of relatedness within and among groups. Like other members of the Egernia group, L. kintorei has a complex social structure. Generally, I found that L. kintorei at Newhaven live in groups of related individuals, although there is considerable variation among groups in their size, structure and stability across years: some groups involved male–female pairs together for a single season, while one group contained a mated pair and offspring from three separate seasons. It is possible that L. kintorei social groups are not so constrained by habitat availability because of their ability to construct their own shelter sites. However, the importance of stable food sources in the unpredictable

! i! ! ! desert environment could maintain group living in this species because tolerance among group members would allow them to share such resources rather than compete among kin.

Through broader-scale genetic analyses, I found evidence of male-biased dispersal, with no genetic structure among males, but significantly high relatedness among females at short distance classes. This was supported by the recapture data for males, showing a higher level of movement among burrows than for females.

Males were captured, on average, at a higher number of distinct burrows than were females, and the distance between capture points was also higher. There was also sexual head-size dimorphism in this species, with males having longer and broader heads than females, relative to body size. This trait may be maintained in L. kintorei by sexual selection through agonistic encounters among interspecific males.

An analysis of genetic differentiation among localities inhabited by L. kintorei across its distribution was also carried out. Utilising mitochondrial and microsatellite loci, we genotyped samples collected from six localities, revealing high levels of differentiation among them. The extent of this differentiation suggests that regions containing this Threatened species should be managed separately, particularly the southeastern locality of Uluru, and this distinctiveness should be recognised if intervention such as translocation or captive breeding is to be undertaken.

! ii! ! ! Declaration

The work described in this thesis was undertaken in the Conservation Genetics

Laboratory at Macquarie University. I certify that all work described herein is original and has not been submitted, in any form, for a higher degree at any other institution.

Project design, analysis and manuscript writing was undertaken primarily by myself, with guidance from my supervisors, A. Stow and M. Whiting. I, accompanied by field assistants, conducted all fieldwork and sample collection at Newhaven, and

Steve McAlpin provided samples from other localities outlined in Chapter 2.

D. Chapple (Monash University) designed four novel microsatellite loci used and published in this thesis. Trialing and optimisation of these new markers were undertaken by me, and I carried out all other laboratory work described herein. All other contributions are appropriately acknowledged in each chapter.

All animals were handled in accordance with the Macquarie University Animal

Ethics Committee under protocol numbers ARA2011/037 and ARA2012/037. Sample collection was licensed by the Northern Territory Government (NRETAS; permit number 41144).

Siobhan Dennison

March 2015

! iii! ! ! Personal acknowledgements

I feel so lucky to have had so many amazing people by my side along this whole journey (and I apologise in advance for the lengthy thank yous). It’s been a bumpy ride, but I’ve had so much fun along the way, and I’ve not just learned a lot about lizards, but about myself too. Nothing I can say can express how grateful I am to have been surrounded by such incredible people with whom I shared these experiences, and who were always around to catch me when I stumbled a bit. Firstly, to my supervisors, Adam Stow and Martin Whiting. Thank you both so much for giving me the opportunity to work with you, and for the opportunity to do this PhD. I’ve had an amazing time with you both – from fiery adventures in the field to antics in the lab – your guidance and friendship at every twist and turn is something I’ll always be so grateful for. Adam, your support through this whole thing, both academically and personally, has been amazing. Thank you so much for all of your advice, help in the lab and with analyses, for guiding me, and always helping me structure my thoughts. Your door was always open, and your ability to talk me through an existential crisis – reminding me to step back and look at the big picture (or to take a break!) – really helped me see this through. Thank you. Martin, you have no idea how much the words “You can call me any day, anytime” mean to a panicked PhD student – thank you so much for your endless support, encouragement, guidance, and again, always having your door open for a chat (academic or otherwise!), and a good laugh – and for bringing me into the Lizard Lab family! Andy Beattie and Chris Turnbull. You guys have been such amazing mentors and special friends to me. Thank you for your unwavering support, guidance and insights, and some amazing experiences in the field. Josef Schofield and Danae Moore were such amazing hosts at Newhaven, I’m so grateful for all the help you so readily gave, above and beyond anything I could have expected. Your knowledge and love of that country and everything in it really opened up my eyes to that amazing place, and I’m counting the days until I can come back. Steve McAlpin - likewise, thank you for sharing your knowledge of great desert skinks and central Australia (as well as all of your stories and music!), and for always showing a keen interest in my work and progress. I want to thank the Ngalia Warlpiri People, traditional owners of Newhaven lands, and the people at Nyirripi community who shared their knowledge and always made me

! iv! ! ! feel welcome on my visits. Thank you also to the Australian Wildlife Conservancy for allowing me to work at their Newhaven property, particularly Alex James for support. My lab, the Conservation Genetics Lab, is a force to be reckoned with. Our informative and often colourful (or downright dark) conversations over a daily coffee are something that I will always remember with a smile. Shannon Smith, Heather Baldwin, Maria Asmyhr, Jake Coates, Paolo Momigliano, Vince Repaci, Stephen Hoggard and Miranda Christopher – I love you guys! Thanks for teaching me so much, and always keeping me smiling (or belly laughing). Sarah Collison and Liette Vandine - thank you both so much for your endless support and friendship. Sarah – thank you for the tough love – your unconditional support and strength have really helped me push through these last months. Liette – your positive attitude and enthusiasm is infectious – thanks for swooping in when I needed a lift, and being more certain than I that this would get finished! Jono Davis, my brother from another mother, thank you for your love and support – in particular for my ‘assertiveness training’ and showing me that I’m tougher than I thought I was. Finally, to my incredible family – especially my parents, John and Mary. It was you two who first instilled in me a deep love and appreciation of the natural world around me, and encouraged me to always ask questions and delve deeper. Thank you so much for your unwavering and unconditional love, support, and patience throughout this whole journey – I really couldn’t have done it without you. My beautiful siblings, Alyx and Matt, and cousin Danni – you guys have always inspired me to push that little bit harder and go that little bit further no matter what might try to hold you back. I might not admit it all the time, but you all inspire me so much, and make me feel ok to venture out of my comfort zone, knowing you’re always there to pick me up and dust me off. These have been simultaneously the most challenging, rewarding, humbling and (dare I say it) fun years of my life yet – I can’t wait to see what comes next!

! v! ! !

! vi! ! ! Presentations of research from this thesis

Conference presentations

• Dennison S., Whiting M.J., and Stow A.J. (2013) Group composition and mating

system of a social, burrowing lizard. Oral presentation. Australian Society of

Herpetologists. Point Wolstoncroft, NSW, 2013.

• Dennison S., McAlpin S., Whiting M.J. and Stow A.J. (2012) Characterising dispersal

and group structure in a social, burrowing lizard. Oral presentation. World Congress

of Herpetology. Vancouver, Canada, 2012.

• Dennison S., Whiting M.J., and Stow A.J. (2011) Factors influencing group structure

in a social, burrowing lizard. Oral presentation. Australian Evolution Society.

Townsville, QLD, 2011.

• Dennison S., McAlpin S., Whiting M.J. and Stow A.J. (2011) Lizards building a home

for the family. Poster presentation. Genetics Society of Australasia. Melbourne, VIC,

2011.

Invited seminars

• “Group structure in a social, burrowing lizard: the great desert skink, Liopholis

kintorei.” Faculty of Biology, University of Valencia, Spain, 2013.

Media appearances

• PBS Nature miniseries: “Animal Homes” – segment on Great Desert Skinks

• “Science 202” series on 107.3FM – 2SER Breakfast radio, “Social Life of Skinks”

!

! vii! ! !

List of Figures

Fig. 2.1. Distribution and sampling localities of L. kintorei...... 24

Fig. 2.2. Haplotype network constructed from 72 Liopholis kintorei mitochondrial ND4 sequences...... 29

Fig. 2.3. STRUCTURE bar plot showing population assignment of Liopholis kintorei individuals captured from four locations across their distribution...... 33

Fig. 2.4. Discriminant analysis of principal components (DAPC) for K = 2 genotypic clusters of Liopholis kintorei sampled from four locations across their distribution....34

Fig. S2.2.1. STRUCTURE bar plot of K = 4 genotypic clusters of Liopholis kintorei using location information...... 50

Fig. S2.2.2. Output of discriminant analysis of principal components (DAPC) for K =

4 genotypic clusters of Liopholis kintorei, using location information...... 51

Fig. 3.1. Schematic of female and male Liopholis kintorei burrow use at the main study site...... 68

Fig. 3.2. Maximum distance between captures for adult L. kintorei...... 69

Fig. 3.3. Spatial genetic structure of (a) female, (b) male and (c) juvenile Liopholis kintorei sampled at Newhaven...... 72

Fig. 4.1. Average pairwise relatedness between group members pooled across years...... 107

Fig. 4.2. Heat map illustrating the differences in distance to termite mounds, % ground cover and % shrub cover for burrow and control sites...... 108

Fig. S4.1.1. a) Specific growth rates for a given; and b) logistic growth curve for snout-vent length in Liopholis kintorei...... 127

Fig. S4.2.1. Average pairwise relatedness between group members pooled across years when all individuals at the site were included in analyses...... 136

! viii! ! ! Fig. 5.1. Slight positive trend for the extent of SDHS to increase with within-group R in species of the Egernia group...... 148

Fig. 5.2. No relationship was evident between the extent of SDHS and levels of polygyny in species of the Egernia group...... 148

Fig. 5.3. Extent of sexual head size dimorphism (± SE) in species categorised as: non-territorial, specific site defense, territorial...... 154

Fig. A1. Map of active Liopholis kintorei burrow systems located at the main study site...... 205

Fig. A2. Histogram of lizard captures and the proportion of within-season recaptures for each successful trapping night throughout the study...... 206

Fig. A3. Histogram of the cumulative proportion of recaptures per successful trapping night over all three years of study...... 207

Fig. A4. Histograms of the number of captures per individual lizard throughout the study for (a) females, (b) males, and (c) juveniles...... 208

! ix! ! ! List of Tables

Table 2.1. Sample sizes and diversity indices for Liopholis kintorei...... 30

Table 2.2. Locus by locus results for Hardy-Weinberg Equilibrium tests for L. kintorei sample localities...... 31

Table 2.3. Pairwise genetic differentiation between four Liopholis kintorei sampling localities, generated from 585 bp of the mitochondrial ND4 gene...... 32

Table 2.4. Range of Jukes-Cantor genetic distances between four Liopholis kintorei sampling localities calculated from ND4...... 32

Table S2.1.1. Summary statistics and characteristics of four polymorphic microsatellite loci in Liopholis kintorei...... 48

Table S2.3.1. Monthly average maximum temperature (Tmax), average minimum

(Tmin) temperature and average rainfall for Sangster’s Bore, NT...... 52

Table S2.3.2. Monthly average maximum temperature (Tmax), average minimum

(Tmin) temperature and average rainfall for Uluru, NT...... 53

Table S2.3.3. Results of independent t-tests comparing average maximum temperature (Tmax), average minimum (Tmin) temperature and average rainfall at

Sangster’s Bore and Uluru, NT from 1997–2013...... 54

Table 3.1. Species within the Egernia group for which sex-biased dispersal information is available...... 61

Table 3.2. Total number of captures, maximum distance between captures, and number of distinct tunnel systems from which individual L. kintorei were captured at

Newhaven, NT...... 68

Table 3.3. Summary statistics of 8 microsatellite loci used in analyses of Liopholis kintorei...... 70

! x! ! ! Table S3.1. Total number of captures, maximum distance between captures, and number of distinct tunnel systems from which individual L. kintorei were captured at

Newhaven, NT. Only individuals captured >10 times included in analyses...... 86

Table 4.1. Number of Liopholis kintorei individuals of each age and sex class captured in each year of sampling...... 99

Table 4.2. Composition of Liopholis kintorei groups in 2011–2013...... 104

Table 4.3. Average relatedness (R) between Liopholis kintorei within and among social groups at Newhaven each year...... 106

Table 4.4. Environmental correlates of burrow complexes vs. control quadrats...... 109

Table 4.5. Environmental correlates of burrows containing groups vs. solitary individuals...... 109

Table S4.1.1. Growth parameters and predicted SVL (mm) at a given age for

Liopholis kintorei...... 128

Table S4.2.1. Composition of Liopholis kintorei groups at Newhaven in 2011–2013 when all individuals captured were included in analyses...... 131

Table S4.2.2. Average relatedness between Liopholis kintorei within and among social groups at Newhaven each year when all individuals captured were included in analyses...... 135

Table 5.1. Summary of body and head measurements (mm) for male and female L. kintorei...... 147

Table 5.2. Data collected for Egernia group species: the extent of SDHS, within- group relatedness, degree of polygyny (% males that multiply mated), and references are given for each species...... 147

Table 5.3. Species list, data and references used to compare sexual head size dimorphism and territoriality in lizards...... 150

Table 5.4. Phylogenetic generalised least-squares model and parameter estimates

! xi! ! ! for the prediction of the extent of SDHS based on level of territoriality...... 154

Table S5.1. Pearson correlation coefficients for head width, head length and head height in Liopholis kintorei...... 173

Table A1. Morphometric data for all Liopholis kintorei individuals captured at the main site in 2011...... 196

Table A2. Morphometric data for all Liopholis kintorei individuals captured at the main site in 2012...... 197

Table A3. Morphometric data for all Liopholis kintorei individuals captured at sites other than the main site 2012...... 200

Table A4. Morphometric data for all Liopholis kintorei individuals captured at the main site in 2013...... 202

Table A5. GPS waypoints for all burrows at which lizards were captured throughout the study...... 203

Table A6. Burrow occupancy – list of all captures throughout the study...... 209

Table A7. Characteristics of burrow systems and control quadrats...... 218

Table A8. Genotypes for all Liopholis kintorei samples...... 220

Table A9. Variable sites for 17 haplotypes sampled from six localities across the distribution of Liopholis kintorei...... 228

! xii! ! !

! ! !

! ! ! CHAPTER 1

General introduction

Photo by Adam Stow

! 1! ! ! Social organisation

Social organisation is highly diverse across the animal kingdom. It can range from solitary living (Sandell, 1989), through seasonal aggregations during breeding or overwintering periods (Graves and Duvall, 1995), to living in stable, cooperative groups (Cockburn, 1998; Hughes, 1998) or eusocial colonies (Jarvis, 1981; Queller and Strassmann, 1998). Living in kin-based groups in particular has raised many questions and generated extensive discussion since Darwin (1859) first acknowledged the challenge it presented to his Theory of Evolution. The question itself of why animals live in groups, is an extremely important one because the factors that potentially lead to and maintain group living are intimately linked to other key evolutionary processes, including sexual selection (West-Eberhard, 1983), kin selection (Hamilton, 1964), cooperative behaviour (Cockburn, 1998) and conflict

(Emlen, 1982b). Understanding what might drive the evolution and maintenance of group living will aid in our understanding too of these broader evolutionary processes.

Living in kin-based groups is a result of offspring delaying their dispersal to remain with their parents, in some cases, forgoing their own reproduction in order to do so (Emlen, 1982a, 1994). This longer-term association among family members provides the foundation for more complex social behaviours to come about, including cooperative and altruistic behaviours directed toward kin (Hamilton, 1964; Trivers,

1971; Cockburn, 1998; Clutton-Brock, 2002).

The potential costs associated with group living are well established, and include competition for resources among kin (Griffin and West, 2002), increased risk of inbreeding (Storz, 1999), increased conspicuousness to predators (Alexander,

1974) and transmission of parasites (Altizer et al., 2003). Despite these costs, sociality is virtually ubiquitous across the animal kingdom, and researchers continue

! 2! ! ! to explore an increasingly broad array of traits and behaviours that may contribute to its evolution and maintenance.

In vertebrates, most explanations for the evolution of kin-based group living have focused on inclusive fitness benefits, where individuals may maximise their fitness by contributing to the increased survival of their kin (kin-selection; Hamilton,

1964), and ecological constraints, where limited availability of suitable habitat or habitat saturation can mean that the costs of dispersing are sufficiently high that individuals remain with their natal group (Emlen, 1982). A lack of available habitat options might mean that offspring delay dispersal and share resources, while kin selection may influence the tolerance of these individuals in the parental territory.

More recently, some authors have incorporated the benefits of philopatry, and prolonged association with parents and the benefits of group living, rather than the costs and benefits of delayed dispersal per se (e.g. through parental care, increased survival, or access to, or inheritance of, resources or territories; Stacey and Ligon,

1991; Komdeur, 1992). This may, in turn, enable the development of more complex social bonds and cooperative behaviours.

Social organisation in reptiles

Most of what we know about the evolution of vertebrate sociality has come from extensive research on birds and mammals (e.g. Faulkes et al., 1997; Cockburn,

1998; Clutton-Brock et al., 2000; Koenig and Dickinson, 2002). Reptiles, on the other hand, are largely underrepresented in the literature because until relatively recently they were thought to exhibit little, if any, complex social organisation such as parental care or kin-based group living (Doody et al., 2013). However, a fuller understanding of the evolutionary drivers of such behaviours can be achieved through studying their independent origins in taxonomically diverse groups. Reptiles therefore offer an

! 3! ! ! exciting opportunity by which to extend our understanding of the evolutionary transitions to more complex social behaviour. Reptile mating systems range from monogamy to polygynandry, and although generally less complex than those seen in many birds and mammals, they display a broad range of social systems (solitary individuals through to family groups; Chapple, 2003; Uller and Olsson, 2008).

Furthermore, parental care in reptiles is generally limited to tolerance of juveniles in the natal territory, or defense of eggs or offspring (Shine, 1988; Doody et al., 2013).

Many reptiles aggregate because grouping may offer reproductive and thermoregulatory benefits (Graves and Duvall, 1995; Davis Rabosky et al., 2012), or because optimal shelter sites may be a limiting resource in their environment

(Mouton, 2011). Most aggregations in squamates, however, are usually ephemeral, forming during breeding times, or overwintering (Graves and Duvall, 1995), and in many of these cases social bonds are likely to be weak or absent.

Longer-term social associations may occur independently of kinship. For example, in territorial Australian eastern water dragons, females form relatively strong social bonds with males, however they do seek out extra-pair matings.

Female–female social bonds in this species are stronger than those between males and females, and this is associated more with home range overlap than with relatedness (Strickland et al., 2014).

An increasing number of reptiles, however, have been found to exhibit even more complex social organisation, including long-term monogamy and preferentially kin-based interactions (Bull, 2000; Greene et al., 2002; Chapple, 2003; Davis, 2011;

Clark et al., 2012). For example, in the Egernia group of lizards, several species exhibit a level of sociality rarely seen in reptiles, with individuals living in groups containing adults with one or more cohorts of their offspring (Gardner et al., 2001;

OConnor and Shine, 2003; Stow and Sunnucks, 2004a; Chapple and Keogh, 2006).

! 4! ! ! One of the difficulties with studying kin-based social living in reptiles is the cryptic nature of much of the behaviour and communication in many species (e.g. many lizards use chemosensory rather than auditory or visual cues to communicate;

Doody et al., 2013). Even in group-living species, parent–offspring relationships as well as those among other relatives are difficult to ascertain in the wild, and these associations have often only been uncovered through long-term observational studies and the use of molecular techniques (e.g. Gardner et al., 2001; Stow and

Sunnucks, 2004a; Chapple and Keogh, 2005; While et al., 2009a; McAlpin et al.,

2011). In particular, the use of molecular techniques has allowed more comprehensive assessments into the relative importance of kin association in maintaining group living in reptiles. By combining molecular techniques with available life history information, as well as environmental and behavioural data, we can gain a more comprehensive understanding of factors driving sociality in reptiles.

The Egernia group as a model system

One radiation of skinks from Australia, the Egernia group, comprising six genera of scincid lizards (Egernia, Bellatorias, Cyclodomorphus, Liopholis, Lisolepsis, Tiliqua;

Gardner et al., 2008; Pyron et al., 2013) has been adopted as a model for the early evolution of sociality because it comprises species encompassing a range of social complexity, from solitary species (Liopholis inornata, Daniel, 1998; Egernia coventryi,

Clemann et al., 2004; Tiliqua adelaidensis, Fenner and Bull, 2010), through monogamous pairs (Tiliqua rugosa, Bull, 2000) to stable, kin-based groups (Egernia stokesii, Gardner et al., 2001; Egernia saxatilis, OConnor and Shine, 2003; Egernia cunninghami, Stow and Sunnucks, 2004; Liopholis whitii, Chapple and Keogh,

2006). This spectrum is also exhibited intraspecifically, with varying levels of social

! 5! ! ! complexity demonstrated within (While et al., 2009b) and among populations

(Bustard, 1970; Bonnett, 1999) of the same species.

Generally, social groupings in Egernia group species are characterised by long-term pair bonds and delayed juvenile dispersal (Chapple, 2003). Molecular studies over the last decade have confirmed that these groupings comprise ‘family’! groups, often containing and an adult pair and multiple cohorts of full siblings

(Gardner et al., 2001; OConnor and Shine, 2003; Stow and Sunnucks, 2004;

Chapple and Keogh, 2005; McAlpin et al., 2011). Although parental care in Egernia group species is basic compared to the full provisioning seen in many birds and mammals (Clutton-Brock, 1991), these species do exhibit a higher level of care than most other reptile species. In addition to having access to resources within their parents’!territory, juveniles have been shown to experience reduced aggression from unrelated conspecifics in the presence of a parent, suggesting social benefits to philopatry and prolonged parent–offspring associations (OConnor and Shine, 2004;

While et al., 2009a).

Long-term associations with kin, particularly when combined with relatively low levels of dispersal, present a problem for group-living species and increase the risk of inbreeding (Pusey and Wolf, 1996). Several mechanisms have been identified within the Egernia group that may help circumvent this, including recognition of kin (Main and Bull, 1996; Bull et al., 2000; Bull et al., 2001; OConnor and Shine, 2006) and disassortative mating based on relatedness (Gardner et al., 2001; Stow and

Sunnucks, 2004b; Chapple and Keogh, 2005). Although, generally, extra-pair paternity is low in the Egernia lineage (Stow and Sunnucks, 2004a ; Chapple and

Keogh, 2005), individuals may gain benefits in terms of offspring fitness by multiply mating. For example, multiple mating has been shown to increase the heterozygosity of offspring in Liopholis whitii when population viscosity constrains individuals to mate

! 6! ! ! with genetically similar partners (While et al., 2014). Finally, sex-biased dispersal may serve to prevent inbreeding by separating siblings before sexual maturity.

Evidence for sex-biased dispersal in Egernia group species has been mixed. For example, in Egernia major, females tend to move more, and over a wider range, than males (Osterwalder et al., 2004); in E. cunninghami, there was no sex-biased dispersal in naturally vegetated habitats, but in cleared areas it was male-biased

(Stow et al., 2001); in Tiliqua rugosa there was no evidence of sex-biased dispersal

(Bull and Cooper, 1999); while studies in Liopholis whitii (Chapple and Keogh, 2005) and Egernia stokesii yielded ambiguous results (Gardner et al., 2001; Gardner et al.,

2012). Whether males or females are more likely to disperse is thought to be intimately linked to the predominant mating system of a species, with polygynous species showing male-biased, and monogamous species, female-biased dispersal. It is possible that the lack of consistent patterns of sex-biased dispersal in Egernia group species may be the result of inter- and intraspecific variation in mating systems

(i.e. resulting in opposing dispersal patterns; Chapple and Keogh, 2005).

This variable nature of mating and social systems within the Egernia group may reflect facultative responses to local ecological or demographic conditions (Stow et al., 2001; Michael et al., 2010). The emergence of social living is likely a result of constraints imposed by the availability of suitable habitat or resources in the environment (Faulkes et al., 1997; Johnson et al., 2002; Mouton, 2011). From here, it is likely that kin-based group living evolved because of inclusive benefits associated with sharing such resources with relatives (Hamilton, 1964). The presence of social aggregations in the Egernia group has been attributed to constraints such as this

(OConnor and Shine, 2003; Chapple and Keogh, 2005; Michael et al., 2010; but see

Gardner et al., 2007).

! 7! ! ! In order to facilitate between-species comparisons and identify common factors influencing the evolution of group living, it is crucial to expand such studies to include other lineages within the Egernia group from a range of habitats having different constraints.

The majority of Egernia group species that have been studied to date live in mesic regions, and rely on pre-existing shelter sites such as rock crevices or tree hollows (Stow et al., 2001; Duffield and Bull, 2002; OConnor and Shine, 2003;

Duckett et al., 2012), with the exception of Liopholis whitii, which is a burrowing species that still lives in close association with rocky habitats (Chapple, 2005; While et al., 2009b). The great desert skink, Liopholis kintorei, is a burrowing lizard that lives in sandy desert habitat and whose group-living behaviour may not be so constrained by retreat site availability, though suitable habitat may still remain a limiting factor. Furthermore, the unpredictable nature of the desert environment compared with more mesic regions means that factors such as the availability of food resources and increased risks associated with dispersal, may be particularly important factors maintaining kin-based social living in L. kintorei (such as in African mole-rats; Faulkes et al., 1997). However, the ubiquity of sociality across Egernia suggests that this trait is plesiomorphic to the group. Given that other social Egernia group species inhabit more mesic environments, ecological constraints imposed by the desert environment may not be the only factor maintaining kin-based social living.

Investigating members of the Egernia group inhabiting a range of environments will aid in identifying the relative importance of factors that drive and maintain sociality, including resource availability, intraspecific interactions and kin-selection.

McAlpin et al. (2011), using genetic data, revealed that L. kintorei excavates extensive burrow systems inhabited by groups containing adults and juveniles of both sexes. Of the group-living Egernia species, it has the highest level of polygyny (40%

! 8! ! ! of males mate multiply; McAlpin et al., 2011), however the presence of full-siblings from multiple cohorts of offspring suggests that mating pairs stay together across years, as in other Egernia group species.

Thesis overview

This thesis comprises a series of separate but interrelated studies that together will strengthen the knowledge base for using the Egernia group as a model system for the evolution of sociality, and provide important information on the broad- and fine- scale genetic structure of L. kintorei to inform targeted management for the conservation of this Threatened species in central Australia.

This study had two main aims: (1) to document the social organisation of L. kintorei at the Australian Wildlife Conservancy’s Newhaven Sanctuary, approximately

300 km north of the population previously studied by McAlpin (2011). In addition to using molecular tools, as did McAlpin et al. (2011), I use a mark-recapture approach to assess burrow occupancy and group membership across three consecutive years; and (2) to describe aspects of the biology of this elusive species, about which little is known, and discuss these in the broader context of mating systems and group living, with comments on the implications for the species’!conservation. The logistical challenges associated with the remoteness of deserts, make studies within these regions difficult. As a result, deserts are among the most understudied terrestrial biomes globally (Martin et al., 2012). The Australian deserts harbour some of the richest squamate diversity in the world (Pianka and Vitt, 2003), and yet relatively little is still known about the biology of many species in this region. The great desert skink is a nationally listed Threatened species, as well as a culturally important species to traditional Aboriginal groups (McAlpin, 2001; Pearson et al., 2001), making its conservation a high priority for land managers in central Australia. Altered fire

! 9! ! ! regimes and the introduction of the red fox and domestic cat are key factors that have led to the species’!decline, as well as habitat decay from feral herbivores

(Morton, 1990; Cogger et al., 1993; McAlpin, 2001; IUCN, 2014). Knowledge of the biology, genetic structure, dispersal and behavioural characteristics of L. kintorei will contribute to more effective conservation management practices for this species.

This thesis comprises four data chapters addressing the key research objectives described above. Following this general introduction (Chapter 1), Chapter 2 provides a broad scale characterisation of genetic structure among different localities containing L. kintorei across its geographic range in the central arid zone of Australia.

The aim was to identify any regions where the genetic distinctiveness of L. kintorei heightens the conservation value and influences the management options, and the results are discussed in this context. Chapter 3 combines genetic and mark- recapture data to characterise fine-scale movement patterns of L. kintorei among their burrow systems. Chapter 4 describes the social organisation of L. kintorei groups, again using molecular and mark-recapture data. In this chapter I also investigate whether environmental features may predict the placement of burrows in the landscape by carrying out surveys of vegetation cover and the presence of termite mounds: a valuable food resource in the unpredictable desert environment. In

Chapter 5, I investigate sexual size dimorphism in relation to sexual selection in lizards. The chapter starts with a comparison of sexual dimorphism relative to levels of polygyny in Egernia group species, and then extends the investigation across a broader taxonomic range. Finally, Chapter 6 provides a synthesis of the results from the previous chapters, discussing the findings in a broader evolutionary and conservation context, and suggests future directions for research.

Chapter 2 has been published in the journal, PLOS ONE, and Chapters 3, 4 and

5 are in preparation for submission to relevant scientific journals. Because each data

! 10! ! ! chapter is written as a stand-alone piece of work for publication, there is some repetition among them, particularly in reference to sample collection and species descriptions. Likewise, the data used in some chapters may overlap, as they are used to address different aspects of the species’!biology and social organisation.

Sample size may differ among chapters because data were not always available for all individuals for all targeted traits.

Literature cited

Alexander, R. D. 1974. The evolution of social behavior. Annual Review of Ecology

and Systematics 5:325–383.

Altizer, S., Nunn, C., Thrall, P., Gittleman, J., Antonovics, J., Cunningham, A.,

Dobson, A. P., Ezenwa, V., Jones, K., Pedersen, A., Poss, M., and J. Pulliam.

2003. Social organisation and parasite risk in mammals: integrating theory and

emprical studies. Annual Review of Ecology, Evolution, and Systematics

34:517–547.

Bonnett, M. 1999. The ecology, behaviour and genetic relationships of a population

of Egernia striolata. Honours Thesis, Flinders University, Adelaide, Australia.

Bull, C. M. 2000. Monogamy in lizards. Behavioural Processes 51:7–20.

Bull, C. M., and S. J. Cooper. 1999. Relatedness and avoidance of inbreeding in the

lizard, Tiliqua rugosa. Behavioral Ecology and Sociobiology 46:367–372.

Bull, C. M., Griffin, C. L., Lanham, E. J., and G. R. Johnston. 2000. Recognition of

pheromones from group members in a gregarious lizard, Egernia stokesii.

Journal of Herpetology 34:92–99.

Bull, C. M., Griffin, C. L., Bonnett, M., Gardner, M. G., and S. J. Cooper. 2001.

Discrimination between related and unrelated individuals in the Australian lizard

Egernia striolata. Behavioral Ecology and Sociobiology 50:173–179.

! 11! ! ! Bustard, H. 1970. A population study of the scincid lizard Egernia striolata in northern

New South Wales. Kungliga Nederlandse Akademie Wetenschappen 73:186–

213.

Chapple, D. G. 2003. Ecology, life-history, and behavior in the Australian Scincid

genus Egernia, with comments on the evolution of complex sociality in lizards.

Herpetological Monographs 17:145–180.

Chapple, D. G. 2005. Life history and reproductive ecology of White’s skink, Egernia

whitii. Australian Journal of Zoology 53:353–360.

Chapple, D. G., and J. S. Keogh. 2005. Complex mating system and dispersal

patterns in a social lizard, Egernia whitii. Molecular Ecology 14:1215–1227.

Chapple, D. G., and J. S. Keogh. 2006. Group structure and stability in social

aggregations of White’s skink, Egernia whitii. Ethology 112:247–257.

Clark, R. W., Brown, W. S., Stechert, R., and H. W. Greene. 2012. Cryptic sociality in

rattlesnakes (Crotalus horridus) detected by kinship analysis. Biology Letters

8:523–525.

Clemann, N., Chapple, D. G., and J. Wainer. 2004. Sexual dimorphism, diet, and

reproduction in the swamp skink, Egernia coventryi. Journal of Herpetology

38:461–467.

Clutton-Brock, T. 1991. The Evolution of Parental Care. Princeton University Press,

Princeton, New Jersey.

Clutton-Brock, T. H. 2002. Breeding together: kin selection and mutualism in

cooperative vertebrates. Science 296: 69–72.

Clutton-Brock, T. H., Brotherton, P. N., O’Riain, M. J., Griffin, A. S., Gaynor, D.,

Sharpe, L., Kransky, R., Manser, M. B., and G. M. McIlrath. 2000. Individual

contributions to babysitting in a cooperative mongoose, Suricata suricatta.

Proceedings of the Royal Society B: Biological Sciences 267:301–305.

! 12! ! ! Cockburn, A. 1998. Evolution of helping behavior in cooperatively breeding birds.

Annual Review of Ecology and Systematics 29:141–177.

Cogger, H., Cameron, E., Sadlier, R., and P. Eggler. 1993. The Action Plan for

Australian Reptiles. Australian Nature Conservation Agency, Canberra,

Australia.

Daniel, M. C. 1998. Aspects of the ecology of Rosen’s desert skink, Egernia inornata,

in the Middleback Ranges, Eyre Peninsula. Department of Environmental

Science and Management. Honours Thesis, University of Adelaide, South

Australia.

Darwin, C. 1859. On the origins of species by means of natural selection. John

Murray, London, UK.

Davis, A. 2011. Kin presence drives philopatry and social aggregation in juvenile

Desert Night Lizards (Xantusia vigilis). Behavioral Ecology 23:18–24.

Davis Rabosky, A., Corl, A., Liwanag, H., Surget-Groba, Y., and B. Sinervo. 2012.

Direct fitness correlates and thermal consequences of facultative aggregation in

a desert lizard. PLOS ONE 7:e40866.

Doody, J. S., Burghardt, G. M., and V. Dinets. 2013. Breaking the social–non-social

dichotomy: a role for reptiles in vertebrate social behavior research? Ethology

119:1–9.

Duckett, P. E., Morgan, M. H., and A. J. Stow. 2012. Tree-dwelling populations of the

skink Egernia striolata aggregate in groups of close kin. Copeia 2012:130–134.

Duffield, G. A., and C. M. Bull. 2002. Stable social aggregations in an Australian

lizard, Egernia stokesii. Naturwissenschaften 89:424–427.

Emlen, S. T. 1982a. The evolution of helping I. An ecological constraints model. The

American Naturalist 119:29–39.

! 13! ! ! Emlen, S. T. 1982b. The evolution of helping II. The role of behavioural conflict. The

American Naturalist 119:40–53.

Emlen, S. T. 1994. Benefits, constraints and the evolution of the family. Trends in

Ecology and Evolution 9:282–285.

Faulkes CG, Bennett NC, Bruford NW, O’Brien HP, Aguiler GH, Jarvis JUM (1997)

Ecological constraints drive social evolution in the African mole-rats.

Proceedings of the Royal Society of London B, 264, 1619-1627.

Fenner, A. L., and C. M. Bull. 2010. The use of scats as social signals in a solitary,

endangered scincid lizard, Tiliqua adelaidensis. Wildlife Research 37:582–587.

Gardner, M. G., Bull, C. M., Cooper, S. J., and G. A. Duffield. 2001. Genetic evidence

for a family structure in stable social aggregations of the Australian lizard

Egernia stokesii. Molecular Ecology 10:175–183.

Gardner, M. G., Bull, C. M., Fenner, A., Murray, K., and S. C. Donnellan. 2007.

Consistent social structure within aggregations of the Australian lizard, Egernia

stokesii, across seven disconnected rocky outcrops. Journal of Ethology 25:263–

270.

Gardner, M. G., Hugall, A. F., Donnellan, S. C., Hutchinson, M. N., and R. Foster.

2008. Molecular systematics of social skinks: phylogeny and taxonomy of the

Egernia group (Reptilia: Scincidae). Zoological Journal of the Linnean Society

154:781–794.

Gardner, M. G., Godfrey, S. S., Fenner, A. L., Donnellan, S. C., and C. M. Bull. 2012.

Fine-scale spatial structuring as an inbreeding avoidance mechanism in the

social skink Egernia stokesii. Australian Journal of Zoology 60:272–277.

Graves, B. M., and D. Duvall. 1995. Aggregation of squamate reptiles associated

with gestation, oviposition, and parturition. Herpetological Monographs 9:102–

119.

! 14! ! ! Greene, H. W., May, P. G., Hardy, D. L., Sciturro, J. M., and T. M. Farrell. 2002.

Parental behavior by vipers. In: Schuett, G. W., Hoanggren, M., Douglas, M. E.

and Greene, H. W. (Editors) Biology of the Vipers. Eagle Mountain Publishers,

Eagle Mountain, UT.

Griffin, A. S., and S. A. West. 2002. Kin selection: fact and fiction. Trends in Ecology

and Evolution 17:15–21.

Hamilton, W. 1964. The genetical evolution of social behaviour I. Journal of

Theoretical Biology 7:1–16.

Hughes, C. 1998. Integrating molecular techniques with field methods in studies of

social behavior: a revolution results. Ecology 79:383–399.

IUCN. 2014. IUCN Red List of Threatened Species. Version 2014.2.

http://www.iucnredlist.org

Jarvis, J. U. 1981. Eusociality in a mammal: cooperative breeding in naked mole-rat

colonies. Science 212:571–573.

Johnson, D., Kays, R., Blackwell, P., and D. Macdonald. 2002. Does the resource

dispersion hypothesis explain group living? Trends in Ecology and Evolution

17:563–570.

Komdeur, J. 1992. Importance of habitat saturation and territory quality for evolution

of cooperative breeding in the Seychelles warbler. Nature 358:493–495.

Koenig, W., and J. Dickinson. 2002. Ecology and Evolution of Cooperative Breeding

in Birds. Cambridge University Press, Cambridge, UK.

Main, A. R., and C. M. Bull. 1996. Mother–offspring recognition in two Australian

lizards, Tiliqua rugosa and Egernia stokesii. Animal Behaviour 52:193–200.

Martin, L. J., Blossey, B., and E. Ellis. 2012. Mapping where ecologists work: biases

in the global distribution of terrestrial ecological observations. Frontiers in

Ecology and the Environment 10:195–201.

! 15! ! ! McAlpin, S. 2001. A recovery plan for the great desert skink (Egernia kintorei) 2001–

2011. Arid Lands Environment Centre, Alice Springs, Australia.

McAlpin, S. 2011. Social structure and mating system of the Great Desert Skink

Liopholis kintorei. Masters Thesis, Macquarie University, New South Wales.

McAlpin, S., Duckett, P., and A. Stow. 2011. Lizards cooperatively tunnel to construct

a long-term home for family members. PLOS ONE 6:e19041.

Michael, D. R., Cunningham, R. B., and D. B. Lindenmayer. 2010. The social elite:

habitat heterogeneity, complexity and quality in granite inselbergs influence

patterns of aggregation in Egernia striolata (Lygosominae: Scincidae). Austral

Ecology 35:862–870.

Morton, S. R. 1990. The impact of European settlement on the vertebrate animals of

arid Australia: a conceptual model. Proceedings of the Ecological Society of

Australia 16:201–213.

Mouton, P. L. 2011. Aggregation behaviour of lizards in the arid western regions of

South Africa. African Journal of Herpetology 60:155–170.

O’Connor, D., and R. Shine. 2003. Lizards in “nuclear families”: a novel reptilian

social system in Egernia saxatilis (Scincidae). Molecular Ecology 12:743–752.

O’Connor, D., and R. Shine. 2004. Parental care protects against infanticide in the

lizard Egernia saxatilis (Scincidae). Animal Behaviour 68:1361–1369.

O’Connor, D., and R. Shine. 2006. Kin discrimination in the social lizard Egernia

saxatilis (Scincidae). Behavioral Ecology 17:206–211.

Osterwalder, K., Klingenbock, A., and R. Shine. 2004. Field studies on a social lizard:

home range and social organisation in an Australian skink, Egernia major.

Austral Ecology 29:241–249.

! 16! ! ! Pearson, D., Davies, P., Carnegie, N., and J. Ward. 2001. The great desert skink

(Egernia kintorei) in Western Australia: distribution, reproduction and ethno-

zoological observations. Herpetofauna 31:64–68.

Pianka, E. R., and L. J. Vitt. 2003. Lizards: windows to the evolution of diversity.

University of California Press, Berkeley.

Pusey, A., and M. Wolf. 1996. Inbreeding avoidance in animals. Trends in Ecology

and Evolution 11:201–206.

Pyron, R., Burbrink, F., and J. Wiens. 2013. A phylogeny and revised classification of

Squamata, including 4161 species of lizards and snakes. BMC Evolutionary

Biology 13:93–145.

Queller, D. C., and J. E. Strassmann. 1998. Kin selection and social insects.

Bioscience 48:165–175.

Sandell, M. 1989. The mating tactics and spacing patterns of solitary carnivores. In:

Gittleman, J. (Editor). Carnivore Behavior, Ecology, and Evolution. Springer

USA.

Shine, R. 1988. Parental care in reptiles. In: Gans, C. and Huey, R. (Editors). Biology

of the Reptilia Vol. 16. Alan R. Liss, Inc., New York.

Stacey, P., and J. Ligon. 1991. The benefits-of-philopatry hypothesis for the evolution

of cooperative breeding: variation in territory quality and group size effects. The

American Naturalist 137:831-846.

Storz, J. 1999. Genetic consequences of mammalian social structure. Journal of

Mammalogy 80:553–569.

Stow, A. J., and P. Sunnucks. 2004a. High mate and site fidelity in Cunningham’s

skinks (Egernia cunninghami) in natural and fragmented habitat. Molecular

Ecology 13:419–430.

! 17! ! ! Stow, A. J., and P. Sunnucks. 2004b. Inbreeding avoidance in Cunningham’s skinks

(Egernia cunninghami) in natural and fragmented habitat. Molecular Ecology

13:443–447.

Stow, A. J., Sunnucks, P., Briscoe, D. A., and M. G. Gardner. 2001. The impact of

habitat fragmentation on dispersal of Cunningham’s skink (Egernia

cunninghami): evidence from allelic and genotypic analyses of microsatellites.

Molecular Ecology 10:867–878.

Strickland, K., Gardiner, R., Schultz, A. J., and C.H. Frere. 2014. The social life of

eastern water dragons: sex differences, spatial overlap and genetic relatedness.

Animal Behaviour 97:53–61.

Trivers, R. L. 1971. Evolution of reciprocal altruism. Quarterly Review of Biology

46:35–57.

Uller, T., and M. Olsson. 2008. Multiple paternity in reptiles: patterns and processes.

Molecular Ecology 17:2566–2580.

West-Eberhard, M. J., 1983. Sexual selection, social competition, and speciation.

Quarterly Review of Biology 58:155–183.

While, G. M., Uller, T., and E. Wapstra. 2009a. Family conflict and the evolution of

sociality in reptiles. Behavioral Ecology 20:245–250.

While, G. M., Uller, T., and E. Wapstra. 2009b. Within-population variation in social

strategies characterize the social and mating system of an Australian lizard,

Egernia whitii. Austral Ecology 34:938–949.

While, G. M., Uller, T., Bordogna, G., and E. Wapstra. 2014. Promiscuity resolves

constraints on social mate choice imposed by population viscosity. Molecular

Ecology 23:721–732.

! ! !

! 18! ! ! CHAPTER 2

Genetic divergence among regions containing the Vulnerable great desert skink (Liopholis kintorei) in the Australian arid zone

Photo by Christine Turnbull

The following manuscript has been published in the journal PLOS ONE. It is presented in the format required for this journal with the exception of table and figure numbers, and references. These have been altered to read sequentially and consistently throughout this thesis. The published version of this chapter can be found in the Appendix of this thesis.

! 19! ! ! Abstract

Knowledge of genetic structure and patterns of connectivity is valuable for implementation of effective conservation management. The arid zone of Australia contains a rich biodiversity, however this has come under threat due to activities such as altered fire regimes, grazing and the introduction of feral herbivores and predators. Suitable habitats for many species can be separated by vast distances, and despite an apparent lack of current geographical barriers to dispersal, habitat specialisation, exhibited by many desert species, may limit connectivity throughout this expansive region. We characterised the genetic structure and differentiation of the great desert skink (Liopholis kintorei), which has a patchy, but widespread distribution in the western region of the Australian arid zone. As a species of cultural importance to local Aboriginal groups and nationally listed as Vulnerable, it is a conservation priority for numerous land managers in central Australia. Analysis of mitochondrial ND4 sequence data and ten nuclear microsatellite loci across six sampling localities through the distribution of L. kintorei revealed considerable differentiation among sites, with mitochondrial FST and microsatellite F!ST ranging

0.047–0.938 and 0.257–0.440, respectively. The extent of differentiation suggests three main regions that should be managed separately, in particular the southeastern locality of Uluru. Current genetic delineation of these regions should be maintained if future intervention such as translocation or captive breeding is to be undertaken.

! 20! ! ! Introduction

The Australian arid zone occupies 70% of the continent’s landmass and supports an extraordinary biodiversity, including among the world’s richest assemblages of lizards

(Pianka and Vitt 2003; Wilson and Swan 2010). Despite a longstanding recognition of the conservation value of this region, relatively few studies have described patterns of genetic structuring across whole species distributions (Byrne et al. 2008). The remoteness and inaccessibility of central arid Australia makes intensive study of this region logistically very difficult. Characterisation of genetic structure across a landscape is valuable to inform conservation because genetically discrete regions may be under different pressures and require separate management approaches.

Connectivity among these regions may also not correspond with natural boundaries as expected based on observed environmental or geographic features (Barr et al.

2008; Gauffre et al. 2008; Dennison et al. 2012). Moreover, human land use is rapidly changing arid Australia and is posing a number of threats to the biodiversity of the region. Habitat destruction through land clearing, accelerated soil erosion, unsustainable cattle grazing and altered fire regimes continues to threaten inland

Australian biota, as do increased weed and feral animal populations (Burbidge and

McKenzie 1989; Morton 1990; Read and Bowen 2001). Knowledge of levels of genetic connectivity can be used to evaluate localised impacts and prioritise conservation activities.

Although the arid regions of Australia currently lack substantial topographic barriers or expansive waterways, genetic variation may be structured by environmental features. Many reptile species are habitat specialists, restricted to specific habitat types scattered throughout the desert region (Pianka 1969; Pianka

1972). Vast expanses of contiguous habitat types such as dunefields or gibber plain may result in patches of suitable habitat for many species being isolated from each

! 21! ! ! other by hundreds of kilometres, a distance in excess of their likely dispersal capacity

(Pianka 1972; Chapple et al. 2004). For example, Chapple et al. (2004) found considerable phylogeographic structure within Egernia inornata (now Liopholis inornata; Gardner et al. 2008), an arid-Australian lizard, with a number of clades occurring in particular habitat types. Thus, habitat heterogeneity, and the associated habitat specialisation of arid zone lizards, might influence connectivity and restrict gene flow between localities. It is therefore our a priori expectation that localised impacts may threaten genetically distinct components of biodiversity.

There are also additional benefits derived from knowledge of connectivity. For example, it is established that parts of a species distribution that experience prolonged isolation may become sufficiently genetically differentiated that they are worthy of separate management (Moritz 1994; Crandall et al. 2000; Frankham et al.

2011). Furthermore, isolation coupled with reduced effective sizes can lower genetic diversity through drift and impinge on the ability to adapt to environmental change. In such cases, translocations among genetically discrete localities may not be a viable conservation strategy owing to the risk of outbreeding depression (Frankham et al.

2011; but see Weeks et al. 2011). Identifying parts of the distribution requiring separate management enables conservation effort to be prioritised and can guide decisions to translocate, restore, or establish breeding programs (Frankham et al.

2010; Frankham et al. 2011).

The great desert skink, Liopholis (formerly Egernia) kintorei, is a species endemic to the arid-zone of Australia, currently listed as ‘Vulnerable’ (IUCN 2014). It is a large scincid lizard that inhabits sand plains, palaeodrainage lines and undulating gravelly downs (McAlpin 2001). Although its range stretches over a vast area of approximately 1.3 million km2, it is known to be patchily distributed, with its presence recorded at fewer than 100 localities (McAlpin 2011). Great desert skinks exhibit

! 22! ! ! limited dispersal (commonly 0–4 km, up to 9 km; McAlpin, 2011; S. Dennison, in prep.), excavating extensive burrow systems in which close kin live and which may be continuously occupied for up to seven years (McAlpin et al., 2011). It is a culturally important species to traditional Aboriginal groups (Pearson et al., 2001; McAlpin, in prep.), and this combined with its threatened status makes its conservation a high priority for land managers in central Australia. Altered fire regimes and the introduction of the red fox and domestic cat are key factors that have led to the species’ decline, as well as habitat decay from feral herbivores (Morton, 1990;

Cogger et al., 1993; McAlpin, 2001; IUCN, 2014).

Areas containing great desert skinks are known to occur in a number of geographically distant regions within declared conservation areas, and in these a number of monitoring and management actions have previously been undertaken:

Uluru–Kata Tjuta National Park (NT), Newhaven Wildlife Sanctuary (NT), Karlamilyi

National Park (WA), Ngaanyatjarra Indigenous Protected Area (IPA; WA), and in the

Watarru IPA within the Anangu Pitjantjatjara Yankunytjatjara Lands (APY; SA). The extent of isolation or genetic differentiation between these regions is unknown, but likely to be high given the apparent disjunct distribution and low dispersal of the species. Here we use mtDNA sequence data (ND4) and ten microsatellite loci to characterise genetic structure and divergence across the range of L. kintorei. We aim to identify any regions where the genetic distinctiveness of L. kintorei heightens the conservation value and influences the management options.

Methods

Sample collection

Eighty-five L. kintorei samples were collected from six locations throughout the distribution of the species (See Figure 2.1): Australian Wildlife Conservancy’s

! 23! ! ! Newhaven Wildlife Sanctuary (hereafter, Newhaven) and Sangster’s Bore in the northeast of their distribution, Petalu-Docker River toward the centre of their distribution (hereafter Docker River), Uluru-Kata-Tjuta National Park to the east

(hereafter Uluru), Warburton to the west within the Ngaanyatjarra IPA and Watarru within the IPA of the APY Lands at the southern extent of the distribution. Samples from Warburton were provided by the Western Australian Museum. Tissue was obtained via tail-tip biopsy and preserved in 90% ethanol.

Liopholis kintorei is a species that exhibits kin-based social living, and high natal philopatry (McAlpin et al., 2011). To ensure that individuals included in our analyses were not highly related, we included only adults captured from separate, distinct burrow clusters.

SB (23)

NH (30)

DR (1) U (30) WB (2) WT (8)

Fig. 2.1 Distribution (shaded; from Wilson and Swan 2013) and sampling localities of L. kintorei:

Sangster’s Bore (SB), Newhaven (NH), Uluru (U), Docker River (DR), Watarru (WT) and Warburton

(WB). Numbers in brackets depict sample sizes

! 24! ! ! Laboratory procedures

Whole genomic DNA was extracted from tissue using a modified salting-out protocol

(Sunnucks and Hales, 1996). For each sample the mitochondrial gene, NADH dehydrogenase subunit 4 (ND4), was targeted because previous work on reptiles has shown useful levels of variation for intraspecific studies, though its mutation rate is slow enough to allow inference of deeper divergence due to long-term isolation

(Chapple et al., 2004; Greaves et al., 2007). In addition, individuals were genotyped at ten microsatellite loci to characterise fine-scale population dynamics and more recent divergence.

Polymerase chain reactions (PCR) for all markers were carried out using a

PTC-100 Thermocycler (MJ Research, Inc.). Mitochondrial ND4 was amplified using previously developed primers (Arevalo et al.,1994). PCRs were in 20 µL volumes, with each containing 50–100 ng of DNA, 4 uL 5x GoTaq Flexi Buffer (Promega), 2 mM MgCl2, 0.2 µM of each dNTP, 0.125 µM of each primer (ND4 and tRNA-leu) and

1 U Taq Polymerase (Promega). Thermocycling began with an initial denaturation for 5 min at 94 ºC, followed by four touchdown cycles with 94 ºC denaturation for 30 sec, annealing temperatures (55 ºC, 53 ºC, 51 ºC, 49 ºC) for 30 sec, and 72 ºC extension for 45 sec. An additional 35 cycles were carried out at an annealing temperature of 47 ºC, followed by a final 72 ºC extension step for 10 min.

Microsatellite PCRs were carried out in 10 µL volumes containing ~50 ng of DNA. A -

29 M13 sequence was added to the 5’ end of each forward primer to allow for the incorporation of a complementary M13 fluorescent-labelled tag, following the protocol of Schuelke (2000). Four tetranucleotide microsatellite loci were developed concurrent to this study (BX6, CKD, FQR, J3F; see Supplementary material 2.1 for microsatellite design). In addition to these, we utilised six previously developed markers (Est1, Est2, Est9, Est12 – Gardner et al., 1999; Ecu2, Ecu3 – Stow, 2002).

! 25! ! ! All microsatellite loci were amplified with identical reaction conditions: 2 uL 5x GoTaq

Flexi Buffer (Promega), 2.5 mM MgCl2, 0.2 µM of each dNTP, 0.02 µM of forward primer, 0.1 µM reverse primer, 0.1 µM of fluoro-labelled tag (FAM, VIC, NED, or

PET) and 1 U Taq Polymerase (Promega). Thermocycling began with an initial denaturation for 3 min at 94 ºC, followed by five touchdown cycles with 94 ºC denaturation for 30 sec, annealing temperatures (60 ºC, 58 ºC, 56 ºC, 54 ºC, 52 ºC) for 30 sec, and 72 ºC extension for 45 s. An additional 35 cycles were carried out at an annealing temperature of 50 ºC, followed by a final 72 ºC extension step for 10 min. PCR products were visualised by electrophoresis on 2% agarose gel. All PCR purification, sequencing and fragment separation was performed by Macrogen

(Korea).

Data analysis

ND4 sequences were checked by eye and aligned with ClustalW, implemented in

MEGA 5.0 (Tamura et al., 2011), and submitted to GenBank (Accession numbers

KM035773–KM035789). DNA sequences were then translated into amino acid sequences using the vertebrate mitochondrial code. No premature stop codons were observed, indicating that all sequences are true mitochondrial copies. Haplotype and nucleotide diversities were calculated in DnaSP (Librado and Rozas, 2009).

A minimum-spanning network of ND4 haplotypes was constructed in TCS 1.21

(Clement et al., 2000). Global and pairwise ΦST, an analogue of FST (Excoffier et al.,1992), were calculated from ND4 haplotypic data in Arlequin v3.5 (Excoffier and

Lischer, 2010) with 1000 permutations.

Microsatellite alleles were visualised and scored using Peak Scanner 1.0

(Applied Biosystems). To ensure amplification and scoring consistency, at least 10% of samples at each locus were independently rerun and genotyped. Summary

! 26! ! ! statistics, including exact tests for Hardy-Weinberg equilibrium (HWE) and linkage disequilibrium (LD) were conducted in GenAlEx 6.4 (Peakall and Smouse, 2006) and

GENEPOP 4.2 (Raymond and Rousset, 1995). Effective population size (Ne) estimates were calculated utilising the approximate Bayesian framework implemented in ONeSAMP v1.2 (Tallmon et al., 2008).

When calculating FST analogues from highly polymorphic data such as microsatellites, within-population variance can often approach the level of the total variance, resulting in very low FST values even when the populations share no alleles

(Hedrick, 2005; Meirmans, 2006). Following Hedrick (2005) and Meirmans (2006), pairwise fixation index values calculated from microsatellite data (hereafter F!ST) were standardised using the program RECODEDATA 0.1 (Meirmans, 2006).

STRUCTURE v.2.3 (Pritchard et al., 2000) analysis was used to assess genotypic clustering and assignment probabilities. We examined values of K = 1-8

(double the number of sample sites included in the analysis), with 10 replicate runs for each, 105 MCMC iterations burn-in and 104 main iterations. Hubisz et al. (2009) developed a new model for STRUCTURE, which allows the use of sample-site information. This is different to the initial models including location priors, in that it adds power to analyses, but can disregard site information when true clustering is uncorrelated with sampling locations. We used the ‘admixture’ model with correlated allele frequencies, and repetitions were run with and without location information. The number of genetic clusters (K) was determined using the ∆K method of Evanno et al.

(2005).

Discriminant analysis of principal components (DAPC) was used to describe the genetic relationship between sampling localities. DAPC is a multivariate analysis that first uses principal components analysis (PCA) to transform data into uncorrelated components. These components are then analysed using a linear

! 27! ! ! discriminant method, minimising within-group variance while maximising among- group variance (Jombart et al., 2010). Furthermore, this analysis does not assume

HWE and LD, which are often violated when working with natural, small and fragmented populations (Jombart et al., 2010).

DAPC was carried out in the R package adegenet (Jombart, 2008), implemented in R 2.12 (R development core team 2013; www.r-project.org), with K selected using the find.clusters function and Bayesian Information Criterion (BIC).

We also ran DAPC using sample locations as groups (K = 4) to assess the differentiation of our sample sites. PCA was performed in R using the dudi.pca function in the package ade4 (Chessel et al., 2004). Missing data were replaced with the mean (the origin of the X and Y axes, as in Horne et al., 2011). Determining the number of principal components (PCs) to retain as predictors for the discriminant analysis requires a balance between the statistical power of more PCs, and the stability of assignments, though there is no strict rule. Retaining too many PCs with respect to sample size can result in over-fitting the data. This trade-off can be assessed using the a.score function in the R package adegenet (Jombart, 2008).

Analyses were carried out retaining a conservative 13 PCs, the optimal number suggested by a.score, given our relatively small dataset.

Results

Summary statistics

Mitochondrial sequence data

Mitochondrial ND4 sequences of 585 bp were successfully amplified from 72 individuals sampled from the six localities. Sequences contained 23 (3.9%) variable sites, of which 18 (3.1%) were parsimony informative, revealing a total of 17 unique haplotypes (Figure 2.2). Haplotype and nucleotide diversity over all samples were

! 28! ! ! 0.908 and 0.009, respectively, and the variance for both was < 0.0002 (Table 2.1). All haplotypes sampled at Uluru were unique to that locality. Newhaven and Sangster’s

Bore shared haplotypes with only each other, and Docker River and Warburton both shared haplotypes with Watarru. One sample from Warburton was unique to that locality, although small sample sizes may account for this. Because of small sample sizes at Docker River and Warburton, these localities were excluded from further population-level analyses.

Fig. 2.2 Haplotype network constructed from 72 Liopholis kintorei mitochondrial ND4 sequences. Each circle represents a unique haplotype, and the number within indicates its frequency

! 29! ! ! Table 2.1. Sample sizes and diversity indices for Liopholis kintorei. Number of samples (n), number of haplotypes (nh), haplotype diversity (h), nucleotide diversity (π) for all individuals sequenced for ND4 at four of the sampling localities (total n = 69). Number of samples (n), average number of alleles per locus (Na), allelic richness (Ra), number of private alleles (Pa), observed (HO) and expected (HE) heterozygosities over ten microsatellite loci (total n = 91). Diversity indices for Warburton and Docker River could not be calculated because of small sample sizes

Locality n (ND4) nh h π n (µsat) Na Ra Pa HO HE

Newhaven 23 6 0.76 0.0063 21 10.1 6.3 9 0.765 0.780

Sangster’s Bore 23 7 0.78 0.0077 23 10.8 7.2 18 0.811 0.811

Watarru 8 2 0.54 0.0010 8 5.4 5.2 7 0.771 0.688

Uluru 15 2 0.34 0.0006 30 10.5 6.6 18 0.777 0.791

Warburton 2 2 2

Docker River 1 1 1

Microsatellite loci

One of the four remaining sample localities was out of HWE (Newhaven, P < 0.05;

Table 2.2), which may be due to a Wahlund effect from some spatial structure

(Wahlund, 1928). Following Holm-Bonferroni sequential correction (Holm, 1979), 10 out of 180 locus x locus tests for LD (45 per sampling locality) were significant, all of which were for different locus pairs. The presence of some LD is unsurprising, given that it can be common in threatened species that are expected to have small effective population sizes (Frankham et al., 2010). Allelic richness of the four sites ranged from 5.2 to 7.2, and the number of private alleles from 7 to 18. Overall FIS ranged from -0.053 to 0.036, with these positive values probably reflecting a spatial

Wahlund effect (Wahlund, 1928). Ne estimates for each locality were: Newhaven (Ne

= 60.4, 95% CI = 38.8 – 119), Sangster’s Bore (Ne = 41.5, 95% CI = 23.4 – 116.9),

Watarru (Ne = 9.3, 95% CI = 5.8 – 15.9), Uluru (Ne = 21.2, 95% CI = 16.7 – 27.6).

! 30! ! ! Estimates of Ne are sensitive to sample size, and as such ours should be treated with caution due to small sample sizes.

Table 2.2. Locus by locus results for Hardy-Weinberg Equilibrium tests for L. kintorei sample localities.

The inbreeding coefficient (FIS) for each locus is given, as well as over all ten loci. Bold values denote significant FIS (P < 0.05)

Locus Newhaven Sangster’s Bore Watarru Uluru

(n = 21) (n = 23) (n = 8) (n = 30)

BX6 0.170 -0.040 -0.140 0.120

CKD 0.045 0.004 0.143 0.158

FQR -0.019 -0.073 -0.167 0.046

J3F -0.078 0.010 -0.077 -0.169

EST1 0.014 0.130 -0.217 -0.007

EST2 0.036 0.120 0.056 0.192

EST9 -0.063 -0.062 0.167 0.077

EST12 0.217 -0.081 -0.287 -0.028

ECU2 0.017 0.106 -0.021 -0.068

ECU3 -0.048 0.047 -0.077 -0.052

Overall 0.036 0.025 -0.053 0.035

! 31! ! ! Genetic differentiation between localities

All comparisons of genetic differentiation, except one (Newhaven–Sangster’s Bore;

PND4 = 0.108), were high and significant (Table 2.3), indicating very low connectivity between localities. For ND4, overall ΦST = 0.50, P < 0.00001. Pairwise population differentiation for both ND4 (FST) and microsatellites (F!ST) was substantial (Table 2.3;

Juke’s Cantor distances between localities are given in Table 2.4). Newhaven and

Sangster’s Bore were the least differentiated from each other, with low and not- significant FST for ND4, though microsatellite differentiation was relatively high. Uluru was the most differentiated from all other localities in all comparisons.

Table 2.3. Pairwise genetic differentiation between four Liopholis kintorei sampling localities, generated from 585 bp of the mitochondrial ND4 gene (FST; lower diagonal) and ten microsatellite loci

(F!ST; upper diagonal). All values in bold are significant (PND4 < 0.0001; Pmsat < 0.05)

Newhaven Sangster’s Bore Watarru Uluru

Newhaven 0.285 0.257 0.302

Sangster’s Bore 0.047 0.380 0.440

Watarru 0.494 0.502 0.263

Uluru 0.627 0.622 0.938

Table 2.4. Range of Jukes-Cantor genetic distances between four Liopholis kintorei sampling localities calculated from ND4 Newhaven Sangster’s Bore Watarru

Newhaven

Sangster’s Bore 0.000-0.016

Watarru 0.007-0.012 0.005-0.014

Uluru 0.007-0.016 0.009-0.017 0.010-0.014

! 32! ! ! STRUCTURE analysis yielded a best-fit value of K = 2 without location information, and K = 4 with location information. When K = 2 was considered, one cluster comprised the Uluru samples, and the other clumped Newhaven, Sangster’s

Bore and Watarru together (Figure 2.3). When K = 4 including location information, there was clear delineation of sample sites with high assignment probabilities. Only plots for K = 2 are given here, but see Supplementary material 2.2 for plots using location priors (K = 4).

A similar result was found in the DAPC: the lowest BIC value was indeed K =

2 (BIC = 124), however the BIC score for K = 4 was negligibly higher (BIC = 126).

Again, when K = 2, Uluru samples were largely separate from the rest (Figure 2.4).

However, using location information in the DAPC showed considerable genetic separation between all four sampling localities, with the largest separation between the first and second, as well as the first and third principal components, lying between Uluru and the other localities (Supplementary material 2.2, Figure S2.2.2).

When considering the first and third principal components, Sangster’s Bore and

Newhaven overlapped, but the other localities remained separate.

Fig. 2.3. STRUCTURE bar plot showing population assignment of Liopholis kintorei individuals captured from four locations across their distribution: Newhaven (NH), Uluru (UL), Watarru (WT) and

Sangster’s Bore (SB)

! 33! ! !

Fig. 2.4. Discriminant analysis of principal components (DAPC) for K = 2 genotypic clusters of

Liopholis kintorei sampled from four locations across their distribution. The y-axis represents the density of individuals along the given discriminant function. One cluster comprised the Uluru samples, and the other clumped Newhaven, Sangster’s Bore and Watarru together

Discussion

We show genetic partitioning among regions containing the Vulnerable lizard

Liopholis kintorei. The level of genetic divergence among regions at ND4 (Jukes-

Cantor distances, Table 2.4) was below that expected among different species of the genus Liopholis (Chapple et al., 2004), but high enough at both nuclear and mtDNA to demonstrate restricted gene flow. Each of the localities contained similar levels of genetic variation and individuals from the Uluru-Kata-Tjuta region were most genetically distinctive (Tables 2.1, 2.3, 2.4). Our estimates of genetic divergence, in addition to environmental differences experienced among regions, indicate that each of these are reservoirs of important genetic variation and point to the risk of outbreeding depression should interbreeding occur.

Outbreeding depression is lowered reproductive fitness in generations subsequent to crossing of individuals from genetically differentiated parts of their distribution (Frankham et al., 2011). A decision tree developed by Frankham et al.

(2011) assists conservation managers to assess the probability of its occurrence, and

! 34! ! ! thus decide whether populations should be kept separate. Applying this to the data for Liopholis kintorei indicates a risk of outbreeding depression should individuals be translocated. The third part of the tree suggests that if sites have been isolated from each other for 500 years or more, there is a high risk of outbreeding depression and they should remain separated. A crude estimate of divergence times based on a commonly cited mitochondrial calibration of 1.3–2% sequence divergence per million years (see Chapple and Keogh, 2004; Gardner et al., 2008), suggests that the Uluru lineage may have split from the others between 350 kya and 1.31 million years ago.

Although estimates based on such a coarse calibration should be treated with caution, these localities have likely been isolated from each other for a period of time in the order of hundreds of thousands of years.

Furthermore, there are environmental differences between the sites that may contribute to localised adaptation, another factor that flags the possibility of outbreeding depression from translocation according to Frankham et al. (2011).

Similarly, Crandall et al. (2001) propose evaluating ‘ecological exchangeability’ along with estimates of genetic divergence to decide on the parts of species distributions to be managed separately. Given the latitudinal range over which L. kintorei is distributed, it is not surprising that there are environmental differences across our sampling localities. For example, mean monthly minimum and maximum temperatures are consistently and significantly cooler at Uluru than at the northernmost locality of Sangster’s Bore (see Supplementary material 2.3, Tables

S2.3.1–3, for climate data and statistical tests). Average annual rainfall at Sangster’s

Bore is 479 mm compared with 320 mm at Uluru (t32 = 2.09, P = 0.045, Tables

S2.3.1–3), though rainfall patterns differ with Uluru receiving more winter rain.

Sangster’s Bore experiences more days above 35 ºC and 40 ºC than Uluru

(respectively, 169.6 vs. 109.3 days ≥ 35 ºC and 52.9 vs. 32.1 days ≥ 40 ºC, Table

! 35! ! ! S2.3.3; climate data from Australian Bureau of Meteorology, www.bom.gov.au; data not available for Watarru). Given their ectothermic physiology, life history traits in reptiles have been demonstrated to be linked to altitudinal or latitudinal variability in climate (Forsman and Shine, 1995; Chapple, 2005; Sunday et al., 2011). As such, the significant climatic differences between our sampling localities are likely to be relevant to L. kintorei, and may have led to adaptive differences in, for example, thermal tolerance or seasonal activity.

The habitat in which L. kintorei is found in these areas varies also. The

Sangster’s Bore and Newhaven localities occur along and adjacent to palaeodrainage lines, with the species’ preferred habitat in semi-saline spinifex plains dominated by soft spinifex (Triodia pungens) and inland tea tree (Melaleuca glomerata). At Watarru, L. kintorei were found within open mulga woodland or

Eremophila and woolybutt (Eragrostis eriopoda) grass shrubland, and at Uluru they occur on sand plains and flat swales dominated by either hard (Triodia basedowi) or soft spinifex (T. pungens and T. schinzii; McAlpin, 2011).

Within other Liopholis species, mitochondrial sequence divergence is generally > 8% (Chapple and Keogh, 2004). Jukes Cantor genetic distances were considerably lower than this in L. kintorei, with less than 2% divergence between localities. Uluru was the most genetically distinct in all analyses, and shared no haplotypes with any other site; Watarru was also highly differentiated from the other areas, but shared haplotypes with other two southwestern localities (Docker River and Warburton). More sampling at Docker River and Warburton is required to determine the true extent of differentiation among these localities. Sangster’s Bore and Newhaven to the north were highly differentiated based on the microsatellite data set (F!ST = 0.285), but were not significantly differentiated at mitochondrial ND4

(FST = 0.047). This discrepancy, taking into account the different inheritance and

! 36! ! ! mutation rates for these genetic markers (Sunnucks, 2000) might indicate a contemporary barrier to dispersal but higher levels of historical gene flow. As a result, we recognise three main delineations among localities for conservation management: one to the north (Sangster’s Bore and Newhaven), one to the southeast (Uluru), and one in the southwest (Watarru, Docker River, Warburton).

Given the isolation of localities implied by an apparently patchy distribution (McAlpin,

2001; McAlpin et al., 2011) and the genetic differentiation among localities investigated here, genetic diversity, if lost, may not be replenished by migration.

Effective population sizes are substantially lower than actual sizes in wild populations, with the ratio between them often approximating 0.1 (Frankham, 1995), and up to 0.5 (Mace and Lande, 1991). Census size estimates for L. kintorei are estimated to be low (<500 at Uluru; McAlpin, 2001), and genetic Ne estimates in this study were all low (138–232, though these estimates should be treated with caution due to small sample sizes). The Vulnerable status of this species and low estimated population sizes suggest that genetic diversity and viability may be eroded rapidly over time. The threatening processes attributed to the decline of great desert skinks have not been removed, and consequently this is likely to continue. If genetic erosion is allowed to proceed, this can render localised parts of the distribution vulnerable to inbreeding and inbreeding depression (Stow and Sunnucks, 2004). Translocations to bring about a so-called ‘genetic rescue’ have been demonstrated to dramatically reverse the effects of inbreeding depression (Madsen et al., 1996; Weeks et al.,

2011). In the case of L. kintorei, unique haplotypes at some localities and high differentiation estimates suggest that if this need eventuates, parts of the distribution selected for translocations need to be carefully chosen to avoid the risk of outbreeding depression.

! 37! ! ! Knowledge of connectivity combined with landscape management of biological processes is needed to conserve biodiversity (Crandall et al., 2000). Conservation management for L. kintorei should prioritise the preservation of suitable habitat, in particular addressing recent and localised changes in fire regimes and predation pressure to reduce the risk of further localised population declines and thus erosion of genetic diversity. While further sampling needs to be conducted at Watarru,

Docker River and Warburton, the evidence suggests three main delineations for management: (1) Uluru to the southeast, (2) Newhaven and Sangster’s Bore to the north, and (3) Watarru, Docker River and Warburton to the southwest. Uluru in particular should be considered separately for management, and this distinctiveness should be recognised if intervention such as translocation or captive breeding is to be undertaken.

Acknowledgements

Thank you to (in alphabetical order) A. Beattie, M. Gillings, A. Jung, C. Turnbull and

M. Whiting for help with sample collection, and to the Western Australian Museum for providing the samples from Warburton. Many thanks to numerous people from

Uluru–Kata Tjuta National Park, particularly N. Tjakalyiri, J. Clayton, D. Walkabout and K. Bennison. Thanks to the community of Watarru who assisted in the project there, particularly F. Young, T. Mervin and M. Pan. Thanks also to the Yuendumu

WoC Rangers for their assistance at Sangster’s Bore and to the Docker River WoC

Rangers for their assistance at Petalu. Thank you to staff at the Australian Wildlife

Conservancy (AWC), in particular to J. Schofield and D. Moore for invaluable support onsite, and A. James for guidance. All animals were handled in accordance with

Macquarie and Charles Darwin University Animal Ethics committee recommendations (ARA 2008/025 and ARA 2011/037), and sample collection

! 38! ! ! licensed by the Northern Territory Parks and Wildlife Commission and the South

Australian Department of Environment and Heritage. K. Miller, K. Smith and M.

Gardner gave assistance with the development of the L. montana microsatellite

primers. This project was funded by Macquarie University (to SD and SM), the Joyce

W. Vickery Fund (to SD), and AWC. DC received funding from Monash University

Faculty of Science Linkage Project Application Support (LPAS) Scheme. P.

Momigliano offered statistical advice and comments on the manuscript, and we also

thank M. Whiting for constructive comments on the manuscript.

Literature cited

Arevalo, E., Davis, S. K., and J. W. Sites (1994) Mitochondrial DNA sequence

divergence and phylogenetic relationships among eight chromosome races of

the Sceloporus grammicus complex (Phrynosomatidae) in central Mexico.

Systematic Biology 43:387–418.

Barr, K. R., Lindsay, D. L., Athrey, G., Lance, R. F., Hayden, T. J., Tweddale, S. A.,

and P. L. Leberg. 2008. Population structure in an endangered songbird:

maintenance of genetic differentiation despite high vagility and significant

population recovery. Molecular Ecology 17:3628–3639.

Burbidge, A. A., and N. L. McKenzie. 1989 Patterns in the modern decline of western

Australia’s vertebrate fauna: causes and conservation implications. Biological

Conservation 50:143–198.

Byrne, M., Yeates, D. K., Joseph, L., Kearney, M., Bowler, J., Williams, M. A., Cooper,

S., Donnellan, S. C., Keogh, J. S., Leys, R., Melville, J., Murphy, D. J., Porch,

N., and K. H. Wyrwoll. 2008. Birth of a biome: insights into the assembly and

! 39! ! ! maintenance of the Australian arid zone biota. Molecular Ecology 17:4398–

4417.

Chapple, D. G. 2005. Life history and reproductive ecology of White’s skink, Egernia

whitii. Australian Journal of Zoology 53:353–360.

Chapple, D. G., and J. S. Keogh. 2004. Parallel adaptive radiations in arid and

temperate Australia: molecular phylogeography and systematics of the Egernia

whitii (Lacertilia: Scincidae) species group. Biological Journal of the Linnean

Society 83:157–173.

Chapple, D. G., Keogh, J. S., and M. N. Hutchinson. 2004. Molecular phylogeography

and systematics of the arid-zone members of the Egernia whitii (Lacertilia:

Scincidae) species group. Molecular Phylogenetics and Evolution 33:549–561.

Chessel, D., Dufour, A., J. Thioulouse. 2004. The ade4 package – I: One-table

methods. R News 4:5–10.

Clement, M., Posada, D., K. A. Crandall. 2000. TCS: A computer program to estimate

gene genealogies. Molecular Ecology 9:1657–1659.

Cogger, H., Cameron, E., Sadlier, R., and P. Eggler. 1993. The Action Plan for

Australian Reptiles. Australian Nature Conservation Agency, Canberra,

Australia.

Crandall, K. A., Bininda-Emonds, O. R., Mace, G. M., and R. K. Wayne. 2000.

Considering evolutionary processes in conservation biology. Trends in Ecology

and Evolution 15:290–295.

Dennison, S., Smith, S. M., A. J. Stow. 2012. Long-distance gene flow and habitat

specificity of the rock-dwelling coppertail skink, Ctenotus taeniolatus. Austral

! 40! ! ! Ecology 37:258–267.

Evanno, G., Regnaut, S., and J. Goudet. 2005. Detecting the number of clusters of

individuals using the software STRUCTURE: A simulation study. Molecular

Ecology 14:2611–2620.

Excoffier, L., and H. E. Lischer. 2010. Arlequin suite ver 3.5: A new series of programs

to perform population genetics analyses under Linux and Windows. Molecular

Ecology Resources 10:564–567.

Excoffier, L., Smouse, P., and J. Quattro. 1992. Analysis of molecular variance

inferred from metric distances among DNA haplotypes: application to human

mitochondrial DNA restriction data. Genetics 131:479–491.

Forsman, A., and R. Shine. 1995. Parallel geographic variation in body shape and

reproductive life history within the Australian scincid lizard Lampropholis

delicata. Functional Ecology 9:818–828.

Frankham, R. 1995. Effective population size/adult population size ratios in wildlife: a

review. Genetical Research 66:95–107.

Frankham, R., Ballou, J. D., and D. A. Briscoe. 2010. Introduction to Conservation

Genetics (2nd Edition). Cambridge University Press, Cambridge, UK.

Frankham, R., Ballou, J. D., Eldridge, M. D., Lacy, R. C., Ralls, K., Dudash, M. R., and

C. B. Fenster. 2011. Predicting the probability of outbreeding depression.

Conservation Biology 25:465–75.

Gardner, M. G., Cooper, S. J., Bull, C. M., and W. N. Grant. 1999. Isolation of

microsatellite loci from a social lizard, Egernia stokesii, using a modified

enrichment procedure. Journal of Heredity 90:301–304.

Gardner, M. G., Hugall, A. F., Donnellan, S. C., Hutchinson, M. N., and R. Foster.

2008. Molecular systematics of social skinks: phylogeny and taxonomy of the

! 41! ! ! Egernia group (Reptilia: Scincidae). Zoological Journal of the Linnean Society

154:781–794.

Gauffre, B., Estoup, A., Bretagnolle, V., and J. F. Cosson. 2008. Spatial genetic

structure of a small rodent in a heterogenous landscape. Molecular Ecology

17:4619–4629.

Greaves, S. N., Chapple, D. G., Gleeson, D. M., Daugherty, C. H., and P. E. Ritchie.

2007. Phylogeography of the spotted skink (Oligosoma lineoocellatum) and

green skink (O. chloronoton) species complex (Lacertilia: Scincidae) in New

Zealand reveals pre-Pleistocene divergence. Molecular Phylogenetics and

Evolution 45:729–739.

Hedrick, P. W. 2005. A standardized genetic differentiation measure. Evolution

59:1633–1638.

Holm, S. 1979. A simple sequentially rejective multiple test procedure. Scandinavian

Journal of Statistics 6:65–70.

Horne, J. B., Momigliano, P., Welch, D. J., Newman, S. J., and L. van Herwerden.

2011. Limited ecological population connectivity suggests low demands on self-

recruitment in a tropical inshore marine fish (Eleutheronema tetradactylum:

Polynemidae). Molecular Ecology 20:2291–306.

Hubisz, M. J., Falush, D., Stephens, M., and J. K. Pritchard. 2009. Inferring weak

population structure with the assistance of sample group information. Molecular

Ecology Resources 9:1322–1332.

IUCN. 2014. IUCN Red List of Threatened Species. Version 2014.1.

http://www.iucnredlist.org

Jombart, T. 2008. adegenet: a R package for the multivariate analysis of genetic

markers. Bioinformatics 24:1403–1405.

! 42! ! ! Jombart, T., Devillard, S., and F. Balloux. 2010. Discriminant analysis of principal

components: a new method for the analysis of genetically structured

populations. BMC Genetics 11:94.

Librado, P., and J. Rozas. 2009. DnaSP v5: A software for comprehensive analysis of

DNA polymorphism data. Bioinformatics 25:1451–1452.

Mace, G. M., and R. Lande. 1991. Assessing extinction threats: toward a reevaluation

of IUCN threatened species categories. Conservation Biology 5:148–157.

Madsen, T., Stille, B., and R. Shine. 1996. Inbreeding depression in an isolated

population of adders Vipera berus. Biological Conservation 75:113–118.

McAlpin, S. 2001. A recovery plan for the great desert skink (Egernia kintorei) 2001–

2011. Arid Lands Environment Centre, Alice Springs, Australia.

McAlpin, S. 2011. Social structure and mating system of the great desert skink

Liopholis kintorei. Masters Thesis, Macquarie University, Sydney, Australia.

McAlpin, S., Duckett, P., and A. Stow. 2011. Lizards cooperatively tunnel to construct

a long-term home for family members. PLOS ONE 6(5):e19041

Meirmans, P. G. 2006. Using the AMOVA framework to estimate a standardized

genetic differentiation measure. Evolution 60:2399–2402.

Morton, S. R. 1990. The impact of European settlement on the vertebrate animals of

arid Australia: a conceptual model. Proceedings of the Ecological Society of

Australia 16:201–213.

Peakall, R., and P. E. Smouse. 2006. GENALEX 6: Genetic analysis in Excel.

Population genetic software for teaching and research. Molecular Ecology

Notes 6:288–295.

Pearson, D., Davies, P., Carnegie, N., and J. Ward. 2001. The great desert skink

(Egernia kintorei) in Western Australia: distribution, reproduction and ethno-

zoological observations. Herpetofauna 31:64–68.

! 43! ! ! Pianka, E. R. 1969. Habitat specificity, speciation and species density in Australian

desert lizards. Ecology 50:498–502.

Pianka, E. R. 1972. Zoogeography and speciation of Australian desert lizards: an

ecological perspective. Copeia 1:127–145.

Pianka, E. R., and L. J. Vitt. 2003. Lizards: windows to the evolution of diversity.

University of California Press, Berkeley.

Pritchard, J. K., Stephens, M., and P. Donnelly. 2000. Inference of population structure

using multilocus genotype data. Genetics 155:945–959.

Raymond, M., and F. Rousset. 1995. GENEPOP (version 1.2): population genetics

software for exact tests and ecumenicism. Journal of Heredity 86:248–249.

Read, J., and Z. Bowen. 2001. Population dynamics, diet and aspects of the biology of

feral cats and foxes in arid South Australia. Wildlife Resources 28:195–203.

Schuelke, M. 2000. An economic method for the fluorescent labeling of PCR

fragments. Nature Biotechnology 18:233–234.

Stow, A. J. 2002. Microsatellite loci from the Cunningham’s skink (Egernia

cunninghami). Molecular Ecology Notes 2:256–257.

Stow, A. J., and P. Sunnucks. 2004. Inbreeding avoidance in Cunningham’s skinks

(Egernia cunninghami) in natural and fragmented habitat. Molecular Ecology

13:443–447.

Sunday, J. M., Bates, A. E., Dulvy, N. K. 2011. Global analysis of thermal tolerance

and latitude in ectotherms. Proceedings of the Royal Society B: Biological

Sciences 278:1823–1830.

Sunnucks, P. 2000. Efficient genetic markers for population biology. Trends in Ecology

and Evolution 15:199–203.

! 44! ! ! Sunnucks, P., and D. F. Hales. 1996. Numerous transposed sequences of

mitochondrial cytochrome oxidase I-II in aphids of the genus Sitobion

(Hemiptera: Aphididae). Molecular Biology and Evolution 13:510–524.

Tallmon, D. A., Koyuk, A., Luikart, G., and M. A. Beaumont. 2008. ONESAMP: a

program to estimate effective population size using approximate Bayesian

computation. Molecular Ecology Resources 8:299–301.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., and S. Kumar. 2011.

MEGA5: Molecular evolutionary genetics analysis using maximum likelihood,

evolutionary distance, and maximum parsimony methods. Molecular Biology

and Evolution 28:2731–2739.

Wahlund, S. 1928. Zusammensetzung von populationen und

korrelationserscheinungen vom standpunkt der vererbungslehre aus betrachtet.

Hereditas 11:65–106.

Weeks, A. R., Sgro, C. M., Young, A. G., Frankham, R., Mitchell, N. J., Miller, K. A.,

Byrne, M., Coates, D. J., Eldridge, M. D., Sunnucks, P., Breed, M., James, E.

A., and A. A. Hoffmann. 2011. Assessing the benefits and risks of

translocations in changing environments: a genetic perspective. Evolutionary

Applications 4:709–725.

Wilson, S., and G. Swan. 2010. A Complete Guide to Reptiles of Australia (3rd

Edition). Reed New Holland, Sydney, Australia.

! ! !

! 45! ! ! Supplementary material 2.1

Characterisation of four polymorphic microsatellite loci for the great desert skink, Liopholis kintorei

Genomic DNA was extracted from tissue samples collected from a closely related species, Liopholis montana, using the DNeasy extraction kit (Qiagen) according to the manufacturers protocol. High throughput DNA sequencing was performed with

319ng/µL of DNA on 1/8 of a 70 x 75 PicoTiterPlate using the 454 GS-FLX platform

(Roche Applied Science) at the Australian Genome Research Facility. This technique has been described fully elsewhere (Gardner et al., 2011). A total of 207, 381 contigs were sequenced yielding 1,419 microsatellite loci (di-, tri-, tetra-, penta- and hexanucleotide repeats), identified using MSATCOMMANDER version 0.8 (Faircloth,

2008). Primer 3 (Untergasser et al., 2012) was used to design primer pairs for 187 loci, following the recommendations of Gardner et al. (2011). Ten of these were selected for amplification trials, four of which successfully cross-amplified in Liopholis kintorei and were used in this study (BX6, CKD, FQR, J3F).

Polymerase chain reactions (PCRs) were carried out in 10 µL volumes containing ~50 ng of DNA. A -29 M13 sequence was added to the 5’ end of each forward primer to allow for the incorporation of a complementary M13 fluorescent- labelled tag, following the protocol of Schuelke (2000). Each locus was amplified with identical reaction conditions: 2 uL 5x GoTaq Flexi Buffer (Promega), 2.5 mM MgCl2,

0.2 µM of each dNTP, 0.02 µM of forward primer, 0.1 µM reverse primer, 0.1 µM of fluoro-labelled tag (FAM, VIC, NED, or PET) and 1 U Taq Polymerase (Promega).

Thermocycling began with an initial denaturation for 3 min at 94 ºC, followed by five touchdown cycles with 94 ºC denaturation for 30 sec, annealing temperatures (60 ºC,

58 ºC, 56 ºC, 54 ºC, 52 ºC) for 30 sec, and 72 ºC extension for 45 s. An additional 35

! 46! ! ! cycles were carried out at an annealing temperature of 50 ºC, followed by a final 72

ºC extension step for 10 min. PCR products were visualised by electrophoresis on

2% agarose gel. All PCR purification, sequencing and fragment separation was performed by Macrogen (Korea). Microsatellite alleles were visualised and scored using Peak Scanner 1.0 (Applied Biosystems). To ensure amplification and scoring consistency, at least 10% of samples at each locus were independently rerun and genotyped.

Markers were tested on a panel of 50 Liopholis kintorei individuals captured at the Australian Wildlife Conservancy’s Newhaven Sanctuary, NT, Australia. Tests, including exact tests for Hardy-Weinberg Equilibrium (HWE) and linkage disequilibrium (LD) were conducted in GenAlEx 6.4 (Peakall and Smouse, 2006) and

GENEPOP 4.2 (Raymond and Rousset, 1995) with Holm-Bonferroni corrections for multiple tests (Holm, 1979).

Four tetranucleotide microsatellite loci (BX6, CKD, FQR, J3F) successfully amplified in L. kintorei, with the number of alleles at each locus ranging from 4 to 13.

No locus showed any significant deviation from HWE and no pairs of loci were found to be in significant linkage disequilibrium after Holm-Bonferroni correction (Table

S2.1.1).

! 47! ! ! Table S2.1.1. Summary statistics and characteristics of four polymorphic microsatellite loci in Liopholis kintorei. Number of samples (n), number of alleles per locus (Na), observed (HO) and expected (HE) heterozygosities and the inbreeding coefficient (FIS). None of the loci showed significant deviation from

HWE after Holm-Bonferroni correction

Name Primer sequence Motif Size (bp) n Na HO HE FIS

F: TGAGAACCACTGAGCCACAGG BX6 (ATCT)13 120-172 50 10 0.720 0.827 0.139 R: ACAGGCAAGTAATAGGCATGAGATAGA

F: TGCACCCAATGCACTGACAATGA CKD (CTTT)15 206-254 49 13 0.796 0.878 0.103 R: GACTACAAGCCTCTTAGGAGCAGGA F: TGGTAAATGGCTTTGGGGCCATAC FQR (ACAG)10 110-122 50 4 0.800 0.707 -0.122 R: CGAAGTCTAGAGTTGCGAATGGTAGGT

F: TGTGATTGGTTTCCCTGATTCAAA J3F (CTGT)11 144-172 50 5 0.360 0.453 0.214 R: TGAGTCAAGATGGAGGACGATGG

Literature cited

Faircloth, B. C. 2008. MSATCOMMANDER: Detection of microsatellite repeat arrays

and automated, locus-specific primer design. Molecular Ecology Resources

8:92–94.

Gardner, M. G., Fitch, A. J., Bertozzi, T., and A. J. Lowe. 2011. Rise of the machines

– recommendations for ecologists when using next generation sequencing for

microsatellite development. Molecular Ecology Resources 11:1093–1101.

Holm, S. 1979. A simple sequentially rejective multiple test procedure. Scandinavian

Journal of Statistics 6:65–70.

Peakall, R., and P. E. Smouse. 2006. GENALEX 6: Genetic analysis in Excel.

Population genetic software for teaching and research. Molecular Ecology Notes

6:288–295.

Raymond, M., and F. Rousset. 1995 GENEPOP (version 1.2): population genetics

software for exact tests and ecumenicism. Journal of Heredity 86:248–249.

! 48! ! ! Schuelke, M. 2000. An economic method for the fluorescent labeling of PCR

fragments. Nature Biotechnology 18:233–234.

Untergasser, A., Cutcutache, I., Koressaar, T., Ye, J., Faircloth, B. C., Remm, M.,

and S. G. Rozen. 2012. Primer3-new capabilities and interfaces. Nucleic Acids

Research 40:e115.

! ! !

! 49! ! ! Supplementary material 2.2

Genetic divergence plots utilising location information priors

Genetic substructure of Liopholis kintorei sampled from four locations across their distribution: Newhaven, Uluru, Watarru and Sangster’s Bore. Plots show the results of STRUCTURE analysis and discriminant analysis of principal components (DAPC) when a location information prior is used. Four clusters correspond with the four sample localities.

Fig. S2.2.1. STRUCTURE bar plot of K = 4 genotypic clusters of Liopholis kintorei. ‘Admixture’ model was used with correlated allele frequencies, and a location information prior. The letters on the x-axis refer to the sampling localities: Newhaven (NH), Uluru (UL), Watarru (WT) and Sangster’s Bore (SB)

! 50! ! !

Fig. S2.2.2. Output of discriminant analysis of principal components (DAPC) for K = 4 genotypic clusters of Liopholis kintorei, using a location information prior. Each point represents an individual genotype, each sample site being depicted by a different colour and encompassed by a 95% confidence ellipse. Eigenvalue plots represent the amount of genetic variation contained in each principal component. The X and Y axes in (a) are the first and second principal components, respectively, and in (b) the first and third

Literature cited

Jombart, T. 2008. adegenet: a R package for the multivariate analysis of genetic

markers. Bioinformatics 24:1403–1405.

Jombart, T., Devillard, S., and F. Balloux. 2010. Discriminant analysis of principal

components: a new method for the analysis of genetically structured

populations. BMC Genetics 11:94.

Pritchard, J. K., Stephens, M., and P. Donnelly. 2000. Inference of population

structure using multilocus genotype data. Genetics 155:945–959.

! 51! ! !

Supplementary material 2.3

Climate data and analysis for northern and southern regions containing Liopholis kintorei

(Data from the Australian Bureau of Meteorology – http://www.bom.com.au)

Table S2.3.1. Monthly average maximum temperature (Tmax), average minimum (Tmin) temperature and average rainfall for Sangster’s Bore (station no.

015666, Rabbit Flat), Northern Territory. Data shown are averaged over the last 17 years of data (1997-2013), and standard errors are given in parentheses

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC ANNUAL

52 38.81 37.28 35.96 33.92 28.24 25.46 26.08 29.54 34.49 37.14 38.51 38.36 33.65

! Mean Tmax (0.44) (0.47) (0.58) (0.46) (0.81) (0.35) (0.28) (0.38) (0.39) (0.44) (0.34) (0.40) (0.24)

24.08 23.48 21.72 16.74 11.74 7.68 6.86 8.43 14.51 18.47 21.95 23.52 16.59 Mean Tmin (0.28) (0.20) (0.36) (0.49) (0.52) (0.56) (0.65) (0.44) (0.51) (0.45) (0.33) (0.16) (0.24)

105.05 98.14 79.26 22.43 14.60 7.34 4.78 1.42 6.18 21.37 42.69 76.15 479.41 Rainfall (22.24) (19.38) (28.02) (8.72) (5.62) (2.85) (2.79) (0.98) (3.01) (5.08) (8.47) (11.16) (58.89)

Days ≥ 35 ºC! 25.9 20.6 21.0 11.3 0.9 0.0 0.0 1.7 15.9 23.5 24.4 24.4 169.6

Days ≥ 40 ºC! 12.6 7.3 2.7 0.7 0.0 0.0 0.0 0.0 0.6 7.6 11.3 10.1 52.9

! standard errors could not be calculated, as data for individual years were not available for these variables. Data averaged from 1996-2013

! ! ! ! !

Table S2.3.2. Monthly average maximum temperature (Tmax), average minimum (Tmin) temperature and average rainfall for Uluru (station no. 015635, Yulara

Airport), Northern Territory. Data shown are averaged over the last 17 years of data (1997–2013), and standard errors are given in parentheses

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC ANNUAL

38.68 36.65 33.62 29.83 24.36 20.09 20.35 23.82 29.29 32.15 34.35 35.95 29.94 Mean Tmax (0.39) (0.50) (0.51) (0.51) (0.50) (0.28) (0.41) (0.46) (0.53) (0.56) (0.51) (0.39) (0.22)

23.22 22.53 19.35 14.65 9.26 5.21 4.54 6.12 11.09 15.04 18.46 20.97 14.20 Mean Tmin (0.39) (0.39) (0.49) (0.37) (0.51) (0.45) (0.41) (0.34) (0.33) (0.41) (0.35) (0.30) (0.20)

28.22 51.21 40.69 15.07 10.34 20.95 19.53 5.45 8.27 26.35 47.81 46.11 320.01 Rainfall

53 (5.23) (20.70) (13.60) (6.55) (4.66) (9.21) (6.77) (1.75) (4.19) (8.80) (7.53) (12.67) (48.57) ! Days ≥ 35 ºC! 25.3 19.3 14.3 2.9 0.1 0.0 0.0 0.0 3.4 9.8 14.3 19.9 109.3

Days ≥ 40 ºC! 11.8 6.9 1.7 0.0 0.0 0.0 0.0 0.0 0.0 0.8 4.2 6.7 32.1

! standard errors could not be calculated, as data for individual years were not available for these variables. Data averaged from 1983–2013!

!

! ! ! ! !

Table S2.3.3. Results of independent t-tests comparing average maximum temperature (Tmax), average minimum (Tmin) temperature and average rainfall at Sangster’s Bore and Uluru, Northern

Territory from 1997–2013. Results are given for the hottest month (January), the coldest month (July), and the annual average

Variable n df t P-value

Tmax January 17 32 0.22 0.414

July 17 32 11.43 <0.001

Annual 17 32 11.36 <0.001

Tmin January 17 32 1.80 0.040

July 17 32 3.01 0.003

Annual 17 32 7.54 <0.001

Rainfall January 17 32 3.36 0.001

July 17 32 -2.01 0.026

Annual 17 32 2.09 0.045

!

! 54! ! ! ! !

CHAPTER 3

Sex-biased dispersal and fine-scale movement in a group-living

lizard, the great desert skink, Liopholis kintorei

Photo by Martin Whiting

The following manuscript is currently in preparation for submission to the journal Copeia. It is presented in the format required for this journal with the exception of table and figure numbers. These have been altered to read sequentially throughout this thesis.

! 55! ! ! ! !

Abstract

In vertebrates, much of our understanding of sex-biased dispersal is from work on mammals and birds. While the factors that influence sex-biased dispersal remain controversial, polygyny appears to be commonly associated with male-biased dispersal and monogamy with either female-biased dispersal, or no sex bias.

Identifying factors that predict which sex disperses will benefit from knowledge gained by studies of additional vertebrate lineages. We used genotypic data from eight microsatellite loci and mark-recapture data to investigate sex-biased movement and dispersal in the great desert skink, Liopholis kintorei, a lizard previously described as polygynous, that lives in groups of related individuals in discrete burrow complexes. High natal-site fidelity for females was inferred by the presence of high and significant pairwise relatedness at short distances, while the level of relatedness between males did not vary from random at any distance class. Immature individuals were also found to show high natal-philopatry. These data were concordant with recapture data that showed that males (N = 15) both moved further, and used a higher number of distinct burrow systems than females (N = 26) and immatures (N =

16). We therefore conclude that dispersal in L. kintorei is male-biased and this corroborates the prediction of male-biased dispersal in polygynous species. Further, this may have implications for conservation management of this Threatened species in central Australia, and we consider our result in this context.

Introduction

Numerous explanations for sex-biased dispersal have been proposed, including competition for resources (Greenwood, 1980), competition for mates (Dobson, 1982)

! 56! ! ! ! !

and inbreeding avoidance (Pusey, 1987). Although these hypotheses are not- mutually exclusive, their relative importance remains debated (Lawson Handley and

Perrin, 2007; Mabry et al., 2013). Whether males or females have a greater propensity to disperse may be associated with their mating tactics, because this would determine the resources for which they compete (Favre et al., 1997; Perrin and Mazalov, 2000; Lawson Handley and Perrin, 2007). In polygynous mating systems, males contribute less to parental investment and more to male–male competition for females, while females may direct more of their resources to defending a breeding territory or rearing young (Lawson Handley and Perrin, 2007).

In these cases, dispersal of immature males reduces competition for mates among male kin, while females benefit from maintaining breeding resources in a familiar natal territory (Favre et al., 1997; Perrin and Mazalov, 2000). This might mean that females benefit from either staying in the natal territory, or dispersing only as far as they need to in order to have access to sufficient resources for survival and reproduction (Stamps, 1977), while males might have to move further afield to find territory that is not yet occupied by dominant males (Johansson et al., 2008).

In socially monogamous systems (e.g. many birds), dispersal is often female- biased (Greenwood, 1980; Perrin and Mazalov, 2000) when breeding resources such as nest sites are limiting (Perrin and Mazalov, 2000), or there is no sex-bias when competition for mates is more or less equal between the sexes (Dobson, 1982; Favre et al., 1997). Despite the general patterns of male-biased dispersal in socially polygynous systems and female-biased dispersal (or no-bias) in monogamous systems (Greenwood, 1980; Dobson, 1982; Perrin and Mazalov, 2000), numerous species exist that do not conform to these patterns (e.g. Williams and Rabenold,

! 57! ! ! ! !

2005; Nagy et al., 2007), which implies that mating systems are not solely responsible for influencing patterns of sex-biased dispersal.

Although the mating system of a species plays a key role in the direction and degree of sex-biased dispersal, this relationship is complicated by processes such as inbreeding avoidance, cooperation and competition among kin (Devillard et al., 2004;

Lawson Handley and Perrin, 2007).

The availability of suitable habitats may also influence the degree of sex- biased dispersal, particularly when the risks of locating and moving to a new territory are high (Favre et al., 1997). Limitation of valuable habitat resources has been suggested to influence population dynamics such as social organisation (Fulgione et al., 2003; Mouton, 2011) and aggression among conspecifics (Langkilde and Shine,

2004), and the relative habitat requirements, and costs to each of the sexes associated with such limitation, may dictate the dispersal characteristics of a species

(Perrin and Mazalov, 2000). If suitable habitat is limited, dispersal may be constrained in one sex more than the other (e.g. Stow et al., 2001). Particularly for social species, kin cooperation is important for the acquisition and maintenance of home sites and resources. The benefits of this cooperation may favour residents over immigrants within a site, particularly in acquiring or defending resources (Le Galliard et al., 2003; Lawson Handley and Perrin, 2007). Social bonds among individuals of one sex are likely to reinforce philopatry in that sex, and contribute to sex-biased dispersal (Lawson Handley and Perrin, 2007).

The majority of studies on sex-biased dispersal have focused on birds and mammals, while reptiles have received much less attention. Reptiles display enormous diversity in many life history traits that might influence dispersal

! 58! ! ! ! !

characteristics, including their mating systems (long-term monogamy to polygynandry

– Bull, 2000; Fitze et al., 2005), their reproductive modes (both oviparity and viviparity; Olsson and Shine, 2003), and social organisation (solitary to stable kin- based family groups; Chapple, 2003). This variation provides a valuable opportunity for understanding the factors that predict sex-biased dispersal across phylogenetically distinct taxa.

The Egernia group (sensu Gardner et al., 2008) is a unique radiation of scincid lizards from Australia that has received much attention in studies of the early evolution of kin-based social living. A number of species within this group live in stable kin-based social groups with high levels of social and genetic monogamy, traits rarely seen in reptiles (Gardner et al., 2002; Chapple, 2003; O’Connor and

Shine, 2003; Stow and Sunnucks, 2004a; Chapple and Keogh, 2005). It has been hypothesised that the evolution of such kin-based social organisation has been influenced by low levels of dispersal relating to habitat limitation (Berry et al., 2005;

Michael et al., 2010; Mouton, 2011), where a lack of available retreat sites means that individuals share, rather than compete for them. For example, in Egernia striolata the formation of social groups is associated with the structure of rocky outcrops (Michael et al., 2010). The inclusive benefits of sharing these limited resources with kin may outweigh the costs of competing for them, however this cost– benefit ratio may be disparate between the sexes.

A key characteristic of lizards in the Egernia group is that they are tightly associated with pre-existing permanent retreat sites (Chapple, 2003) such as rock crevices (E. cunninghami, E. saxatilis, E. stokesii, Liopholis whitii) or tree hollows

(Bellatorias frerei, E. striolata). Often these refuges are irregularly distributed

! 59! ! ! ! !

throughout the environment, and separated by expanses of unsuitable habitat (Stow et al., 2001; Duffield and Bull, 2002; Michael et al., 2010), which is likely to greatly influence the lizards’ dispersal characteristics (Gardner et al., 2001; Stow et al.,

2001).

As observed in the Egernia group, species that live in small groups tend to be characterised by long-term socially monogamous pair bonds (O’Connor and Shine,

2003; Stow and Sunnucks, 2004a; Chapple and Keogh, 2006). However, some polygyny is also common, and levels of extra-pair matings vary both within and across species (Stow and Sunnucks, 2004a; Chapple and Keogh, 2005; While et al.,

2009b).

The intraspecific variation in mating systems within the Egernia group makes it a good system in which to investigate sex-biased dispersal and the relative importance of mating systems and habitat availability. While there are few studies within Egernia explicitly investigating sex-biased dispersal, the results so far have been varied, with no clear patterns emerging between sex-biased dispersal and factors such as mating system, social structure or habitat availability (Table 3.1).

! ! !

! 60! ! ! ! ! ! ! ! ! Table 3.1. Species within the Egernia group for which sex-biased dispersal information is available. No clear pattern is evident between sex-biased dispersal and factors such as mating system, social organisation or habitat constraints Species' Mating'system' Social'organisation' Habitat'! Dispersal' References' (level'of'polygyny)' constraints' Bellatorias*major* low'moderate1! group'living2! moderate3! female'biased4! Klingenbock!et!al.,!2000;!! * Osterwalder!et!al.,!2004! ! Bellatorias*frerei* unknown! unknown! low5! male'biased6! Fuller!et!al.,!2005! * Egernia*cunninghami* low! kin'based!groups! high7! male'biased;!no'bias8! Stow!&!Sunnucks,!2004a;!! * Stow!et!al.,!2001! ! Egernia*stokesii* low! kin'based!groups! high7! female'biased;!no!bias9! Gardner!et!al.,!2001;!! * Gardner!et!al.,!2012! ! Liopholis*whitii* low;!moderate! kin'based!groups! high10! equivocal! Chapple!&!Keogh,!2006;!! * While!et!al.,!2009! 61 ! ! Liopholis*kintorei* low;!high11! kin'based!groups! low'moderate?12! tested!here! McAlpin!et!al.,!2011;!! * Chapter!4,!this!thesis! ! Tiliqua*rugosa* low! unrelated!pairs13! low! no!bias! Bull,!1998,!1999! 1 non-random associations with respect to sex; male–-female pairs common 2 groups observed including young of the year, though no evidence for genetic relatedness 3 specific habitat requirements such as logs as retreat sites and open canopy but are able to take advantage of anthropogenic disturbances 4 spatial study showed that females have larger home range sizes than males (larger range of movement rather than dispersal per se) 5 utilises a wide range of habitat features 6 higher relatedness among females than males within sample sites 7 tightly associated with pre-existing permanent retreat sites (rock crevices) 8 male-biased in cleared habitats, no bias in naturally forested areas 9 female-biased dispersal at the “group” level (Gardner et al., 2001), no bias at a broader scale (“outcrop” level; Gardner et al., 2012) 10 although they excavate burrows, L. whitii are still tightly associated with rocky shelter sites 11 McAlpin et al. (2011) found high polygyny at Uluru, while no evidence for polygyny was found at the Newhaven site (Chapter 4 this thesis) 12 burrowing species that can excavate their own shelter sites, however may still be limited by suitable habitat (Chapter 4 this thesis) 13 solitary outside of the mating season

! ! ! ! ! ! ! One species within the Egernia group, Liopholis kintorei, is unique in that individuals excavate extensive burrow systems in which close kin live, including immature lizards of different age cohorts (McAlpin et al., 2011). As such, unlike other members of the Egernia group that exhibit kin-based group-living, its social organisation may not be as restricted by habitat availability in the landscape, although the availability of high-quality sites for burrow excavation may still be a limiting factor. Despite this, L. kintorei is a group-living species, with some burrows being continuously occupied for up to 7 years and containing offspring from multiple cohorts (McAlpin et al., 2011). Individuals are known to reside with and mate with the same partners across multiple years (McAlpin et al., 2011; Chapter 4 this thesis), though genetic polygyny has been described (40% of males at Uluru were found to have multiply mated with females from different burrows; McAlpin et al., 2011).

In this study, we set out to determine whether L. kintorei, a large, obligate burrowing skink inhabiting the central arid zone of Australia (Cogger, 2014), exhibits sex-biased dispersal. Despite the apparent abundance of suitable habitat throughout its central Australian range, L. kintorei has a very patchy distribution, with its presence being recorded at less than 100 localities (McAlpin, 2011). It is a viviparous species, and the burrows may enable gravid females, as well as newborns, to behaviourally thermoregulate at tunnel entrances and simultaneously shelter from avian predators, particularly crucial at these vulnerable life stages.

The level of genetic polygyny previously described by McAlpin et al. (2011) in

L. kintorei is relatively high within the Egernia group. However, another study

(Chapter 4 this thesis) found no evidence of genetic polygyny in the population under investigation here, although that study identified only seven mated pairs, which may not provide a true representation of the mating system of this population.

Characterising the dispersal of L. kintorei, a species that does not appear to be

! 62! ! ! ! ! ! ! limited by availability of existent retreat sites because it can excavate its own, and appears to show intraspecific variation in levels of polygyny, may provide valuable insight into processes that maintain the variation in social strategies within the group, as well as contribute to our understanding of sex-biased dispersal and the generality of hypotheses put forward to explain it.

We combined molecular analyses of relatedness with mark-recapture data to assess differences in male and female dispersal of L. kintorei, and their relative movement among burrows.

Methods

Study area and field methods

We conducted fieldwork at the Australian Wildlife Conservancy’s Newhaven

Sanctuary (22º 48’ S, 131º 15’ E), approximately 360 km northwest of Alice Springs,

Northern Territory, Australia. Liopholis kintorei burrows in this area occur in semi- saline sandy spinifex grasslands dominated by the hummock grass, Triodia pungens, and the shrub, Melaleuca glomerata. Lizards were trapped at a 20 ha main study site which was intensively sampled during the breeding season across three consecutive years (2011–2013). Here, we refer to discrete complexes of tunnels as ‘burrows’.

Liopholis kintorei burrows at our site extended up to 10 m across with up to 20 tunnel entrances (S. Dennison, unpubl. data). Each year, we mapped the location of any new burrows with a GPS and trapped all active tunnels. Individual tunnels within a burrow complex tend to be directed toward a central point. Because it was difficult to determine the true extent of burrow complexes without excavating the tunnels, all tunnels facing toward the centre of the complex were assumed to be part of that burrow system. We also searched approximately 50–100 m around the extent of the site each year as a buffer to minimise edge effects associated with unidentified

! 63! ! ! ! ! ! ! burrows. Active burrows were recognisable by tracks and the presence of fresh scats in the latrine area. We also trapped lizards at a number of additional locations throughout the Newhaven Sanctuary in 2012 as part of a broader-scale genetic analysis, and include these data in the genetic, but not recapture, analyses of this study.

We captured lizards using Elliott traps baited with tinned peas and corn, set at every tunnel entrance of an active burrow system before dusk each night and checked early each morning (an average of 3064 trap-nights per season). For every lizard caught, we recorded the burrow ID where they were captured, measured snout-vent length (SVL) to the nearest 1 mm with a plastic ruler, and determined sex by hemipene eversion for all mature adults (>160mm; Storr, 1968). A tissue sample was taken from the tip of the tail, and preserved in 90% ethanol. Finally, each lizard was tagged with a subcutaneous passive integrated transponder (PIT-tag).

To more accurately determine fine-scale movements by lizards, we deployed remote PIT-tag readers (Trovan Ltd) in 2012 and 2013. These were set at active burrow systems at the main site for 3–4 nights at a time, with circular aerials positioned around 4 tunnel entrances per burrow, resulting in a total of 453 reading nights.

Molecular methods

DNA was extracted from all lizards, using a modified salting-out protocol (Sunnucks and Hales, 1996). Eight microsatellite loci were used for genetic analyses (EST1,

EST2, EST9 – Gardner et al., 1999; ECU3 – Stow, 2002; BX6, CKD, FQR, J3F –

Dennison et al., 2015; Chapter 2 this thesis). All loci were amplified in 10µL volumes containing ~50 ng of DNA, and to each forward primer, a -29 M13 sequence was added to the 5’ end to allow for the incorporation of a complementary M13

! 64! ! ! ! ! ! ! fluorescent-labelled tag, following the protocol of Schuelke (2000). Polymerase Chain

Reaction (PCR) conditions were identical for each locus: 2 uL 5x GoTaq Flexi Buffer

(Promega), 2.5 mM MgCl2, 0.2 µM of each dNTP, 0.02 µM of forward primer, 0.1 µM reverse primer, 0.1 µM of fluoro-labelled tag (FAM, VIC, NED, or PET) and 1 U Taq

Polymerase (Promega). Thermocycling was carried out using a PTC-100

Thermocycler (MJ Research, Inc.). Initial denaturation was for 3 min at 94 ºC, followed by five touchdown cycles consisting of 30 sec denaturation at 94 ºC, 30 sec annealing (60 ºC, 58 ºC, 56 ºC, 54 ºC, 52 ºC) for 30 sec, and a 45 sec extension step at 72 ºC. A further 35 cycles were then implemented at an annealing temperature of

50 ºC, with a final 72 ºC extension step for 10 min. Successful amplification of PCR products was confirmed by electrophoresis on 2% agarose gel. PCR purification and fragment separation was performed by Macrogen (Korea).

The software, Peak Scanner 1.0 (Applied Biosystems) was used to visualise and score microsatellite alleles. A minimum of 10% of samples were independently re-genotyped to ensure amplification and scoring consistency.

The statistical tests used to explore genetic structure require unlinked loci and an absence of null alleles. To ensure these assumptions were met, we tested for Hardy-

Weinberg equilibrium (HWE) and linkage disequilibrium (LD) on a subset of adults using GenALEx v6.5 (Peakall and Smouse, 2012), GENEPOP v4.2 (Raymond and

Rousset, 1995) and FSTAT v2.9 (Goudet et al., 2002).

Statistical analysis

Assessments of burrow use and recapture distances were only carried out on animals captured at the main study site, and not those in the periphery, because only the main site was intensively sampled across consecutive years. Due to our relatively small dataset, our analyses included all individuals that had been captured ≥ 2 times.

! 65! ! ! ! ! ! ! Analyses were also repeated using a more conservative dataset that included only individuals captured ≥ 10 times, as in Strickland et al. (2014; see Supplementary material 3.1). We recorded the maximum distance between burrows and the number of different burrow systems at which they were encountered, to assess whether there were sex-based differences in movement among burrows and distances traversed.

To assess the spatial genetic structure of L. kintorei, Mantel tests and spatial autocorrelation (SA) analyses were carried out on all adult individuals captured across Newhaven Sanctuary. Mantel tests provide an assessment of isolation by distance, while SA illuminates the shape of the distribution of relatedness by comparing pairwise relatedness of individuals across a range of distance classes.

Dispersal of Liopholis kintorei has previously been estimated to be as high as 4 km

(McAlpin et al., 2011) based on the extent of positive genetic structure. However, estimates such as this based on genetic structure are a reflection of the extent of gene flow across geographic space, and do not directly measure individual dispersal distances. Spatial autocorrelation is an individual-based multilocus analysis that has the power to detect dispersal at a very fine scale (i.e. at the scale of tens to hundreds of metres; Banks and Peakall, 2012). All tests were carried out with 1000 permutations on males, females and immature individuals separately. For the SA analysis, we used both the ‘SinglePops’ and ‘Pops as DClass’ options implemented in GenALEx v6.5 (Peakall and Smouse, 2012). Average pairwise relatedness was calculated for individuals compared at a range of distance classes: 0 m (within- burrows), 50 m, 100 m, 500 m (approximate extent of a sampling locality), 2.5 km and 5 km (comparing individuals across sampling localities). A genetic distance matrix was also constructed to assess the distribution of adult male–female relatedness (excluding male–male and female–female comparisons), analysed in

GenALEx using the ‘SinglePops’ SA option.

! 66! ! ! ! ! ! !

Results

Distance between captures, and burrow use of adult males, females and juveniles

A total of 65 individuals were captured at the main study site (28 females, 17 males,

20 juveniles). This site was resampled in consecutive years, and therefore, individuals captured here were able to be included in analyses of movement among burrows using recapture data. There was a female-skewed adult sex ratio of approximately 0.6:1 in both 2012 and 2013. Although this was not significant (2012 –

14 males, 24 females, χ2 = 2.63, df = 1, P = 0.105; 2013 – 8 males, 14 females, χ2 =

1.64, df = 1, P = 0.201), small sample sizes lower the power of these tests (power =

0.36 and 0.26 in 2012 and 2013, respectively).

Twenty-six females, 15 males and 16 juveniles from the main site were captured ≥ 2 times and used in analyses of recapture distance and burrow use. After correcting for multiple tests, we found no significant difference in the total number of captures between adult males, adult females and juveniles (N = 57, F2,54 = 1.053, P =

0.356, Table 3.2). However, adult males appeared to use more burrow systems than did adult females and juveniles, being captured from a higher number of distinct tunnel systems (N = 57, F2,54 = 7.315, P < 0.005; Table 3.2). Figure 3.1 shows burrow use at the main site for ten representative males and females.

The maximum distance between recaptures for adult males was also greater than for adult females and juveniles (N = 57, F2,54 = 12.732, P < 0.005; Table 3.2 &

Figure 3.2 – juveniles are not included in Figures 3.1 & 3.2, as the majority were only captured from one burrow system). There was no significant difference between adult females and juveniles in either the number of burrows from which they were captured

(P = 0.213) or the maximium distance between captures (P = 0.073).

! 67! ! ! ! ! ! !

Table 3.2. Total number of captures, maximum distance between captures, and number of distinct tunnel systems from which individual L. kintorei were captured at Newhaven, NT

No. captures Max. distance (m) No. tunnel systems N (Mean ± SE) (Mean ± SE) (Mean ± SE)

Male 15 14.80 ± 2.39 83.17 ± 11.42 2.80 ± 0.31

Female 26 15.04 ± 1.63 40.55 ± 8.98 1.96 ± 0.17

Juvenile 16 11.06 ± 2.55 11.22 ± 5.05 1.44 ± 0.13

Females' Males'

0' 50' 100' 150' 200' 250' metres'

Fig. 3.1. Schematic of female (left) and male (right) Liopholis kintorei burrow use at the main study site. Each circle represents a burrow (a complex of tunnels). Each different coloured line represents a single individual’s movement among burrows. For clarity, only ten individuals of each sex are included, and all were captured at least five times over three years of sampling

! 68! ! ! ! ! ! ! a) 200" Females"

150"

100"

max.)distance)(m)) 50"

0" L0101" L0103" L0204" L0501" L0502" L0503" L0504" L0507" L0508" L0601" L0701" L0801" L0902" L1101" L1305" L1402" L1702" L1703" L1801" L1902" L2401" L2404" L2405" L2406" L4301" L4502"

b) 200" Males"

150"

100"

max.)distance)(m)) 50"

0" L0201" L0202" L0702" L0803" L0901" L1304" L1501" L1601" L1602" L1701" L2301" L2402" L3601" L4302" L5201" c) 200" Females"

150"

100"

max.)distance)(m)) 50"

0" 140" 150" 160" 170" 180" 190" 200" 210" 220" SVL)(mm))

d) 200" Males"

150"

100"

max.)distance)(m)) 50"

0" 140" 150" 160" 170" 180" 190" 200" 210" 220" SVL)(mm))

Fig. 3.2. Maximum distance between captures for adult female and male L. kintorei. Histograms show the maximum distance between captures for individual female (a) and male (b) lizards, and scatterplots show the maximum distance between captures for individual females (c) and males (d) of a given SVL at first capture

! 69! ! ! ! ! ! ! Molecular summary statistics

A total of 120 lizards (51 females, 29 males, 40 juveniles) were captured across the whole of Newhaven Sanctuary. All of these individuals were used in genetic spatial analyses, and adults and juveniles were analysed separately. After correcting for multiple tests (Holm, 1979), none of the loci genotyped showed significant linkage disequilibrium or deviation from HWE (Table 3.3).

Table 3.3. Summary statistics of 8 microsatellite loci used in analyses of Liopholis kintorei. Number of individuals genotyped (N), number of alleles (Na), observed heterozygosity (HO), expected heterozygosity (HE), the inbreeding coefficient (FIS) and polymorphic information content (PIC) for each locus are given. None of the loci showed significant deviation from HWE

Locus N! ! Na! HO! HE! FIS! PIC!

BX6! 45! 8! 0.711! 0.758! 0.073! 0.729!

CKD! 45! 8! 0.644! 0.790! 0.195! 0.782!

FQR! 45! 4! 0.711! 0.683! -0.030! 0.640!

J3F! 42! 3! 0.524! 0.503! -0.030! 0.403!

ECU3! 45! 9! 0.867! 0.820! -0.045! 0.788!

EST1! 44! 21! 0.886! 0.915! 0.043! 0.911!

EST2! 42! 14! 0.929! 0.879! -0.044! 0.861!

EST9! 42! 9! 0.905! 0.816! -0.097! 0.791!

! 70! ! ! ! ! ! ! Spatial distribution of relatedness

There was weak, though significant isolation by distance for females (R2 = 0.077, P =

0.001), but not males (R2 = 0.002, P = 0.387). There was no significant relatedness between males and females at any distance class in spatial autocorrelation analyses.

Both females and juveniles captured from the same burrow system were highly related (females – mean relatedness = 0.377, 95% CI = 0.277–0.476, Figure 3.3a; juveniles – mean relatedness = 0.276, 95% CI = 0.173–0.376, Figure 3.3c), and the extent of this positive structure was less than 50 m. Adult males were not significantly related at any distance class (mean relatedness = -0.084, 95% CI = -0.256–0.380;

Figure 3.3b). The results of spatial analyses run on data pooled across all years and from a single year did not differ substantially, nor did the results for the two SA methods used, so the pooled data from the ‘SinglePops’ method is reported here.

The discrepancy between male and female genetic structure is strong evidence for male-biased dispersal (Prugnolle and Meeus, 2002; Banks and Peakall, 2012).

! 71! ! ! ! ! ! !

a 0.600%

0.400%

0.200%

0.000% relatedness) !0.200%

!0.400% 0% 0.05% 0.1% 0.5% 2.5% 5% Distance)Class)(km))

b 0.600%

0.400%

0.200%

0.000% relatedness) !0.200%

!0.400% 0% 0.05% 0.1% 0.5% 2.5% 5% Distance)Class)(km))

c 0.600%

0.400%

0.200%

0.000% relatedness) !0.200%

!0.400% 0% 0.05% 0.1% 0.5% 2.5% 5% Distance)Class)(km))

Fig. 3.3. Spatial genetic structure of (a) female, (b) male and (c) juvenile Liopholis kintorei sampled at

Newhaven, NT. Pairwise relatedness is given across a range of distance bins. Dotted red lines represent the 95% confidence interval around which relatedness is effectively zero – that expected in a randomly mating population

Discussion

We have shown that L. kintorei exhibits male-biased dispersal based on genetic spatial analysis. The spatial distribution of relatedness between adult females and juveniles showed significantly higher levels of relatedness at very short distances, while males showed no genetic structure across the landscape. This discrepancy demonstrates male-biased dispersal, and natal philopatry of females and juveniles

(Banks and Peakall, 2012). In addition, males appear to show a greater level of fine- scale movement than females, with recapture data showing that males not only move further, but also utilise a higher number of burrows, on average, than do females.

! 72! ! ! ! ! ! ! Adult females and juveniles do not differ significantly in their level of movement among burrows.

The adaptive benefits of sex-biased dispersal have been attributed to inbreeding avoidance (separating opposite-sex siblings before sexual maturity), as well as avoiding competition among kin for resources or mates (Perrin and Mazalov,

2000). Male-biased dispersal is predicted to occur in polygynous species

(Greenwood, 1980; Dobson, 1982). Although no polygyny has been observed in L. kintorei at this site, this estimate is based on a small number of mated pairs, and may not be an accurate representation of the mating system. Liopholis kintorei males at

Uluru have been shown to mate multiply with females in several burrow systems in close proximity (McAlpin et al., 2011), and it is still possible that this might be the case at Newhaven as well. The higher level of movement of males among burrows supports the possibility that males might move among burrow systems in search of extra-group females. Male-biased dispersal may not only help prevent siblings from mating, it might also increase inclusive fitness by preventing mate competition among male relatives. Furthermore, by remaining philopatric, female L. kintorei may invest less in searching or competing for refuges, and more in searching for food and in reproduction, while males may move among burrows more, and traverse larger distances to secure matings with multiple females.

Species within the Egernia group that live in stable kin-based groups have generally shown low levels of dispersal (Duffield and Bull, 2002; Chapple and Keogh,

2005), with inbreeding avoidance mostly attributed to disassortative mating based on kin-recognition (Main and Bull, 1996; Bull and Cooper, 1999; Gardner et al., 2001;

Stow and Sunnucks, 2004b; Chapple and Keogh, 2005), or female promiscuity

(While et al., 2014). Our data were not sufficient to test these hypotheses in L.

! 73! ! ! ! ! ! ! kintorei, however, the lack of spatial structure in male–female relatedness suggests that individuals of the opposite sex occupying the same burrow are unrelated.

Delayed dispersal of juveniles, as evidenced by positive fine-scale genetic structure of immature individuals, may increase the likelihood of offspring survival in this species. Liopholis kintorei reach maturity in their 3rd year (S. Dennison, unpubl. data; Chapter 4 this thesis), and in that time they might accrue both direct and indirect benefits from delaying their dispersal, such as access to potentially heritable, high-quality resources (Langkilde et al., 2007), or through protection from predators

(via enhanced vigilance within the group – Lanham and Bull, 2004; or communal retreat sites – McAlpin et al., 2011) or from aggressive conspecific males (While et al., 2009a). In addition to this, though not mutually exclusive, social factors such as kin-based interactions have been demonstrated to influence natal dispersal in lizards.

For example, maternal presence influences sex-biased dispersal of juvenile common lizards (Lacerta vivipara), whereby juvenile females exhibit higher levels of dispersal in the presence of the mother (Le Galliard et al., 2003). Relatedness among conspecifics has also been shown to promote philopatry. Davis (2011) found that juvenile desert night lizards (Xantusia vigilis) are more likely to disperse, and to disperse further, and are less likely to aggregate when released with unrelated individuals, as opposed to those released with genetic relatives.

It appears that the natal philopatry observed in L. kintorei is most likely a result of delayed dispersal, because adult females captured in the final year of sampling had been previously captured as immatures at the same burrow, and the presence of adult sisters in the same burrow (Chapter 4 this thesis) suggests longer-term stability of these social interactions. Delayed dispersal, as well as increasing survival and thus inclusive fitness within the group, may serve to reinforce mother–offspring or sibling bonds, a precursor to kin-based social living.

! 74! ! ! ! ! ! ! Female philopatry is likely to increase the inclusive fitness benefits of joint parental care. Although parental care in Egernia group species is largely restricted to tolerance of young in the natal territory (e.g. While et al., 2009a), adult individuals are more likely to tolerate (and indirectly protect; Sinn et al., 2008; While et al., 2009a) juveniles in their territory if they are more closely related (Emlen, 1995), while there may be conflict directed at juveniles to whom they are not related (While et al.,

2009a). If adult females remain philopatric and share burrows with kin, the average relatedness within the group increases, and conflict should be reduced.

Evidence of sex-biased dispersal in other Egernia is thin. For example,

Gardner et al. (2001) found some evidence of male-biased dispersal in one population of E. stokesii, though in a later investigation of the same species carried out at a broader scale (outcrop-level rather than group-level) found none (Gardner et al., 2012). Chapple and Keogh (2005) found conflicting results in their assessment of sex-biased dispersal in L. whitii, and suggested that this might be a reflection of intraspecific variation in mating systems. The male-biased dispersal and movement in L. kintorei found in this study may be a reflection of polygyny in this species: L. kintorei at Uluru has the highest rate of polygyny recorded in Egernia species

(McAlpin et al., 2011), and as mentioned earlier, although no polygyny has been found among L. kintorei sampled at the current study site (Chapter 4 this thesis), this is based on few identified mated pairs and may not be representative. More intensive tracking at this and replicate sites may shed more light on extra-group social interactions and polygyny in this species.

Dispersal and social organisation may not be constrained in L. kintorei to the extent of that in other Egernia group species due to their burrowing rather than saxicolous lifestyle, although the availability of suitable habitat may still be a limiting factor. However, the tendency for kin-based group living despite potentially lessened

! 75! ! ! ! ! ! ! habitat constraints may also be a result of conserved behaviours within the Egernia lineage (Gardner et al., 2008). Consistency in social organisation (i.e. presence of kin-based groups) among populations of some Egernia species in different-quality habitat lends credence to this suggestion (Gardner et al., 2007; While et al., 2009b).

It is well-known that small, isolated populations are generally more susceptible to extinction (Frankham et al., 2010). Liopholis kintorei is a nationally threatened species in Australia, with a patchy distribution throughout its central-Australian range

(IUCN, 2014). Its decline is attributed to altered fire regimes and feral predators.

Since European settlement, fire suppression has increased the intervals between fires, leading to higher fuel loads. This results in large-scale wildfire events that severely reduce the suitable habitat for many organisms. Moore et al. (in press) showed that L. kintorei burrows have reduced occupancy after fire, and suggested that their risk of predation may be higher because of reduced vegetation cover and increased exposure. If males are more mobile among burrows, as our study suggests, they may be more susceptible to predation than females. Furthermore, in

L. kintorei habitat that has been affected by fire, sex-based differences in dispersal could mean slow natural recolonisation of these areas because of low dispersal among females, and unless viable populations occur nearby, this may be considerably delayed. Management practices for this species should take this into account, and monitor natural recolonisation of areas post-fire, assisting if necessary.

If higher vagility makes males more susceptible to predation, then fire management practice should aim to preserve ground cover to maintain suitable habitat, and provide protection from predators, particularly during the breeding season when males may be most mobile and vulnerable.

That to date there have been no clear patterns evident regarding sex-biased dispersal within the Egernia group (although explicit studies are lacking) may be a

! 76! ! ! ! ! ! ! reflection of the vast intra- and interspecific variation in social organisation, mating systems and habitat requirements within the group. This highlights a unique opportunity to further investigate the relative roles that these factors play in determining the extent and direction of dispersal in a taxon distinct from those that have been most intensively studied (e.g. Greenwood 1980; Le Galliard et al., 2003;

Lawson Handley and Perrin, 2007). We have found male-biased dispersal and female philopatry based on fine-scale genetic analysis in L. kintorei, and determined that males exhibit a higher level of fine-scale movement than females and juveniles at our site. Given the group-living behaviour of this species, male-biased dispersal in this system may serve to reduce inbreeding among close kin, as well as reducing mate competition among male kin. Female philopatry may also reinforce social bonds among kin, as well as increase overall relatedness among group members, thus maximising inclusive fitness within social groups. Although polygyny has not been demonstrated at this study site, it is possible that the higher level of male movement among burrows within our site may reflect males seeking extra-group females for mating opportunities, which has been demonstrated previously from genetic data in this species (McAlpin et al., 2011).

Acknowledgements

We are grateful to everyone who helped with sample collection: M. Asmyhr, A.

Beattie, H. Baldwin, J. Bishton, J. Davis, J. Dennison, M. Gillings, W. Greene, J.

Porter, P. Momigliano and C. Turnbull, and to members of the Nyrripi community for help with tracking burrows. A special thank you to S. McAlpin for help in the field and for sharing his knowledge of great desert skinks in many a wonderful discussion. The land managers at the Australian Wildlife Conservancy’s Newhaven Sanctuary, J.

! 77! ! ! ! ! ! ! Schofield and D. Moore, provided an incredible amount of knowledge and support on site. All animals were handled in accordance with recommendations of Macquarie and Charles Darwin Universities’ animal ethics committees (ARA 2011/037), and sample collection was licensed by the Northern Territory Government (NRETAS permit number 41144). This project was funded by Macquarie University, the Joyce

W. Vickery Fund (to SD), the Rice Memorial Fund (to SD), and the Australian Wildlife

Conservancy.

Literature cited

Banks, S. C., and R. Peakall. 2012. Genetic spatial autocorrelation can readily detect

sex-biased dispersal. Molecular Ecology 21:2092–2105.

Berry, O., Tocher, M. D., Gleeson, D. M., and S. D. Sarre. 2005. Effect of vegetation

matrix on animal dispersal: genetic evidence from a study of endangered skinks.

Conservation Biology 19:855–864.

Bull, C. M. 2000. Monogamy in lizards. Behavioural Processes 51:7–20.

Bull, C. M., and S. J. B. Cooper. 1999. Relatedness and avoidance of inbreeding in

the lizard, Tiliqua rugosa. Behavioral Ecology and Sociobiology 46:367–372.

Chapple, D. G. 2003. Ecology, life-history, and behavior in the Australian Scincid

genus Egernia, with comments on the evolution of complex sociality in lizards.

Herpetological Monographs 17:145–180.

Chapple, D. G., and J. S. Keogh. 2006. Group structure and stability in social

aggregations of white’s skink, Egernia whitii. Ethology 112:247–257.

! 78! ! ! ! ! ! ! Chapple, D. G., and J. S. Keogh. 2005. Complex mating system and dispersal

patterns in a social lizard, Egernia whitii. Molecular Ecology 14:1215–1227.

Cogger, H. 2014. Reptiles and Amphibians of Australia (7th ed.). Collingwood,

Victoria: CSIRO Publishing, Australia.

Davis, A. R. 2011. Kin presence drives philopatry and social aggregation in juvenile

Desert Night Lizards (Xantusia vigilis). Behavioral Ecology 23:18–24.

Devillard, S., Allaine, D., Gaillard, J., and D. Pontier. 2004. Does social complexity

lead to sex-biased dispersal in polygynous mammals? A test on ground-dwelling

sciurids. Behavioural Ecology 15:83–87.

Dobson, F. 1982. Competition for mates and predominant juvenile male dispersal in

mammals. Animal Behaviour 30:1183–1192.

Duffield, G. A., and C. M. Bull. 2002. Stable social aggregations in an Australian

lizard, Egernia stokesii. Naturwissenschaften 89:424–427.

Emlen, S. T. 1995. An evolutionary theory of the family. Proceedings of the National

Academy of Sciences 92:8092–8099.

Favre, L., Balloux, F., Goudet, J., and N. Perrin. 1997. Female-biased dispersal in

the monogamous mammal Crocidura russula: evidence from field data and

microsatellite patterns. Proceedings of the Royal Society B: Biological Sciences

264:127–132.

Fitze, P. S., Le Galliard, J. F., Federici, P., Richard, M., and J. Clobert. 2005. Conflict

over multiple-partner mating between males and females of the polygynandrous

common lizards. Evolution 59:2451–2459.

! 79! ! ! ! ! ! ! Frankham, R., Ballou, J. D., and D. A. Briscoe. 2010. Introduction to Conservation

Genetics (2nd ed.). Cambridge, UK: Cambridge University Press.

Fulgione, D., Rippa, D., De Luca, A., and M. Milone. 2003. Genetic and behavioural

analysis in three typical colonies of Jackdaws (Corvus monedula). Ethology

Ecology and Evolution 15:183–189.

Gardner, M. G., Bull, C. M., and S. J. B. Cooper. 2002. High levels of genetic

monogamy in the group-living Australian lizard Egernia stokesii. Molecular

Ecology 11:1787–1794.

Gardner, M. G., Bull, C. M., Cooper, S. J. B., and G. A. Duffield. 2001. Genetic

evidence for a family structure in stable social aggregations of the Australian

lizard Egernia stokesii. Molecular Ecology 10:175–183.

Gardner, M. G., Bull, C. M., Fenner, A., Murray, K., and S. C. Donnellan. 2007.

Consistent social structure within aggregations of the Australian lizard, Egernia

stokesii across seven disconnected rocky outcrops. Journal of Ethology 25:263–

270.

Gardner, M. G., Cooper, S. J. B., Bull, C. M., and W. N. Grant. 1999. Isolation of

microsatellite loci from a social lizard, Egernia stokesii, using a modified

enrichment procedure. Journal of Heredity 90:301–304.

Gardner, M. G., Godfrey, S. S., Fenner, A. L., Donnellan, S. C., and C. M. Bull. 2012.

Fine-scale spatial structuring as an inbreeding avoidance mechanism in the

social skink Egernia stokesii. Australian Journal of Zoology 60:272–277.

Gardner, M. G., Hugall, A. F., Donnellan, S. C., Hutchinson, M. N., and R. Foster.

2008. Molecular systematics of social skinks: phylogeny and taxonomy of the

! 80! ! ! ! ! ! ! Egernia group (Reptilia: Scincidae). Zoological Journal of the Linnean Society

154:781–794.

Goudet, J., Perrin, N., and P. Waser. 2002. Tests for sex-biased dispersal using bi-

parentally inherited genetic markers. Molecular Ecology 11:1103–1114.

Greenwood, P. J. 1980. Mating systems, philopatry and dispersal in birds and

mammals. Animal Behaviour 28:1140–1162.

Holm, S. (1979). A simple sequentially rejective multiple test procedure.

Scandinavian Journal of Statistics 6:65–70.

IUCN. (2014). IUCN Red List of Threatened Species. Version 2014.2. Retrieved from

www.iucnredlist.org.

Johansson, H., Surget-Groba, Y., and R. S. Thorpe. 2008. Microsatellite data show

evidence for male-biased dispersal in the Caribbean lizard Anolis roquet.

Molecular Ecology 17:4425–4432.

Langkilde, T., O’Connor, D., and R. Shine. 2007. Benefits of parental care: Do

juvenile lizards obtain better-quality habitat by remaining with their parents?

Austral Ecology 32:950–954.

Langkilde, T., and R. Shine. 2004. Competing for crevices: interspecific conflict

influences retreat-site selection in montane lizards. Oecologia 140:684–91.

Lanham, E. J., and C. M. Bull. 2004. Enhanced vigilance in groups in Egernia

stokesii, a lizard with stable social aggregations. Journal of Zoology 263:95–99.

Lawson Handley, L., and N. Perrin. 2007. Advances in our understanding of

mammalian sex-biased dispersal. Molecular Ecology 16:1559–1578.

! 81! ! ! ! ! ! ! Le Galliard, J.-F., Ferrière, R., and J. Clobert. 2003. Mother-offspring interactions

affect natal dispersal in a lizard. Proceedings of the Royal Society B: Biological

Sciences 270:1163–1169.

Mabry, K. E., Shelley, E. L., Davis, K. E., Blumstein, D. T., and D. H. Van Vuren.

2013. Social mating system and sex-biased dispersal in mammals and birds: a

phylogenetic analysis. PLOS ONE 8:e57980.

Main, A. R., and C. M. Bull. 1996. Mother-offspring recognition in two Australian

lizards, Tiliqua rugosa and Egernia stokesii. Animal Behaviour 52:193–200.

McAlpin, S. (2011). Social structure and mating system of the Great Desert Skink,

Liopholis kintorei. Biological Sciences. Macquarie University, North Ryde, NSW,

Australia.

McAlpin, S., Duckett, P., and A. Stow. 2011. Lizards cooperatively tunnel to construct

a long-term home for family members. PLOS ONE 6:e19041.

Michael, D. R., Cunningham, R. B., and D. B. Lindenmayer. 2010. The social elite:

habitat heterogeneity, complexity and quality in granite inselbergs influence

patterns of aggregation in Egernia striolata (Lygosominae: Scincidae). Austral

Ecology 35:862–870.

Mouton, P. L. F. 2011. Aggregation behaviour of lizards in the arid western regions of

South Africa. African Journal of Herpetology 60:155–170.

Nagy, M., Heckel, G., Voigt, C. C., and F. Mayer. 2007. Female-biased dispersal and

patrilocal kin groups in a mammal with resource-defence polygyny.

Proceedings of the Royal Society B: Biological Sciences 274:3019–3025.

! 82! ! ! ! ! ! ! O’Connor, D., and R. Shine. 2003. Lizards in “nuclear families”: a novel reptilian

social system in Egernia saxatilis (Scincidae). Molecular Ecology 12:743–752.

Olsson, M., and R. Shine. 2003. Female-biased natal and breeding dispersal in an

alpine lizard, Niveoscincus microlepidotus. Biological Journal of the Linnean

Society 79:277–283.

Osterwalder, K., Klingenbock, A., and R. Shine. 2004. Field studies on a social lizard:

home range and social organisation in an Australian skink, Egernia major.

Austral Ecology 29:241–249.

Peakall, R., and P. E. Smouse. 2012. GenALEx 6.5: Genetic analysis in Excel.

Population genetic software for teaching and research-an update. Bioinformatics

28:2537–2539.

Perrin, N., and V. Mazalov. 2000. Local competition, inbreeding, and the evolution of

sex-biased dispersal. The American Naturalist 155:116–127.

Prugnolle, F., and D. Meeus. 2002. Inferring sex-biased dispersal from population

genetic tools: a review. Heredity 88:161–165.

Pusey, A. E. 1987. Sex-biased dispersal and inbreeding avoidance in birds and

mammals. Trends in Ecology and Evolution 2:295–299.

Raymond, M., and F. Rousset. 1995. GENEPOP (version 1.2): population genetics

software for exact tests and ecumenicism. Journal of Heredity 86:248–249.

Schuelke, M. 2000. An economic method for the fluorescent labeling of PCR

fragments. Nature Biotechnology 18:233–234.

! 83! ! ! ! ! ! ! Sinn, D.L., While, G.M., and E. Wapstra. 2008. Maternal care in a social lizard: links

between female aggression and offspring fitness. Animal Behaviour 76:1249–

1257.

Storr, G. 1968. Revision of the Egernia whiteii species-group (Lacertilia, Scincidae).

Journal of the Royal Society of Western Australia 51:51–62.

Stow, A. J. 2002. Microsatellite loci from the Cunningham’s Skink (Egernia

cunninghami). Molecular Ecology Notes 2:256–257.

Stow, A. J., and P. Sunnucks. 2004a. High mate and site fidelity in Cunningham’s

skinks (Egernia cunninghami) in natural and fragmented habitat. Molecular

Ecology 13:419–430.

Stow, A. J., and P. Sunnucks .2004b. Inbreeding avoidance in Cunningham’s skinks

(Egernia cunninghami) in natural and fragmented habitat. Molecular Ecology

13:443–447.

Stow, A. J., Sunnucks, P., Briscoe, D. A., and M. G. Gardner. 2001. The impact of

habitat fragmentation on dispersal of Cunningham’s skink (Egernia

cunninghami): evidence from allelic and genotypic analyses of microsatellites.

Molecular Ecology 10:867–878.

Strickland, K., Gardiner, R., Schultz, A. J., and C. H. Frère. 2014. The social life of

eastern water dragons: sex differences, spatial overlap and genetic relatedness.

Animal Behaviour 97:53–61.

Sunnucks, P., and D. F. Hales. 1996. Numerous transposed sequences of

mitochondrial cytochrome oxidase I-II in aphids of the genus Sitobion

(Hemiptera: Aphididae). Molecular Biology and Evolution 13:510–524.

! 84! ! ! ! ! ! ! While, G. M., Uller, T., and E. Wapstra. 2009a. Family conflict and the evolution of

sociality in reptiles. Behavioral Ecology 20:245–250.

While, G. M., Uller, T., and E. Wapstra. 2009b. Within-population variation in social

strategies characterize the social and mating system of an Australian lizard,

Egernia whitii. Austral Ecology 34:938–949.

While, G. M., Uller, T., Bordogna, G., and E. Wapstra. 2014. Promiscuity resolves

constraints on social mate choice imposed by population viscosity. Molecular

Ecology 23:721–732.

Williams, D. A. and K. N. Rabenold. 2005. Male-biased dispersal, female philopatry,

and routes to fitness in a social corvid. Journal of Animal Ecology, 74: 150–159.

! !

! 85! ! ! ! ! ! ! Supplementary material 3.1

Results of analyses of movement among burrows using individuals captured

>10 times

Nineteen females and nine males from the main site were captured ≥ 10 times and included in analyses of recapture distance and burrow use. Only four juveniles were captured ≥ 10 times, and so juveniles were excluded from this analysis.

Using this more conservative dataset, we found no significant difference in the total number of captures between adult males and females (N = 28, F1,26 = 1.156, P =

0.292, Table S3.1). Males appeared to use more burrow systems than females and were captured from a higher number of distinct burrow systems than females (N =

28, F1,26 = 6.751, P = 0.015; Table S3.1). The maximum distance between recaptures for males was also greater than for females (N = 28, F1,26 = 6.588, P = 0.016; Table

S3.1).

Table S3.1. Total number of captures, maximum distance between captures, and number of distinct tunnel systems from which individual L. kintorei were captured at Newhaven, NT

No. captures Max. dist. (m) No. tunnel systems N (Mean ± SE) (Mean ± SE) (Mean ± SE)

Male 9 21.33 ± 1.67 95.03 ± 14.53 3.22 ± 0.43

Female 19 18.74 ± 1.46 45.41 ± 11.36 2.11 ± 0.22

! 86! ! ! ! ! ! ! CHAPTER 4

Variable social organisation in a group-living lizard,

Liopholis kintorei

Photo by Jonathan Davis

The following manuscript is currently in preparation for submission to the Journal of Zoology. It is presented in the format required for this journal with the exception of table and figure numbers. These have been altered to read sequentially throughout this thesis.

! 87! ! ! ! ! ! ! Abstract

Social organisation in vertebrates ranges from solitary individuals to seasonal aggregations, stable kin-based groups or eusocial colonies. Differences in social organisation may be driven by a range of life history, ecological and social factors.

Squamate reptiles exhibit relatively simple social organisation, however one lineage of Australian skinks, the Egernia group, contains a number of species that live in stable social groupings, often comprising highly related individuals. Most species studied within this group inhabit permanent, pre-existing retreat sites, the availability of which has been suggested to delay natal dispersal. However, the great desert skink (Liopholis kintorei) is one species in this group that may not be as limited by retreat-site availability, since it constructs burrows in sandy substrate in central arid

Australia. We investigated the social organisation of L. kintorei using molecular techniques and mark-recapture data collected over three years (2011–2013). We identified 11 social groups, ranging in size from 2 to 9 individuals (mean ± SE = 3.6 ±

0.2), with 23, 40 and 22 individuals being assigned to groups in 2011, 2012 and

2013, respectively. We found high variability among groups in composition, within- group relatedness and stability. Generally, relatedness was higher between individuals within groups (R = 0.31 ±!0.04) than among groups (R = -0.03 ±!0.01).

Fifty-five percent of groups identified had at least two members present across two or more years, however presence of adult offspring suggests that the stability of some groups is even higher. Surveys were also carried out to assess whether environmental features may predict the placement of burrows in the landscape. We found a significant association (F1,39 = 5.09, P = 0.03)!between active burrows and the presence of termite mounds – a stable food resource in an unpredictable desert environment. The variation among groups in social organisation may be a result of

! 88! ! ! ! ! ! ! reduced pressure to find retreat sites, however the requirement of a stable food resource may reduce competition among kin and facilitate group cohesion.!

Introduction

A wide range of social organisation is evident among vertebrates, ranging from solitary individuals (Fenner and Bull, 2010) to temporary aggregations around a common resource (Graves and Duvall, 1995), or from stable, cooperative groups

(including kin-based, or family groups; Hughes, 1998) to eusocial colonies (Jarvis,

1981). The costs of group living are widely accepted and generally universal, including competition for resources (Griffin and West, 2002), increased risk of inbreeding (Storz, 1999), transmission of parasites (Altizer et al., 2003), and increased conspicuousness to predators (Alexander, 1974). The benefits of group living may include enhanced survival through prolonged parental care or access to resources (Dickinson and McGowan, 2005), enhanced vigilance or protection from predators (Stacey, 1986; Lanham and Bull 2004), and increased foraging efficiency

(Blundell et al., 2002).

The relative costs and benefits of group living vary widely among species, influenced by life history (e.g. reproductive mode, age at maturity, delayed dispersal;

Davis et al., 2010) ecological (e.g. habitat availability; Michael, et al., 2010; Mouton,

2011), and social factors (e.g. parental care; While et al., 2009a).

The evolution of vertebrate sociality, particularly living in kin-based social groups, is poorly understood in reptiles (Doody et al., 2013). Most social aggregations in squamates are usually ephemeral, formed during breeding times, or overwintering (Graves and Duvall, 1995), and in many of these cases social bonds are likely to be weak or absent. Until relatively recently, squamate reptiles were thought to exhibit little, if any, complex social behaviour, such as parental care or kin-

! 89! ! ! ! ! ! ! based group living. However, in the past two decades, such behaviours have been uncovered in a range of squamates. For example, juvenile and pregnant timber rattlesnakes aggregate preferentially with kin, while there is no kin structure in other aggregations such as those associated with hibernacula (Clark et al., 2012).

The monophyletic lineage of scincid lizards from Australia, the Egernia group

(comprising the genera Bellatorias, Cyclodomorphus, Egernia, Liopholis, Lissolepsis and Tiliqua; Gardner et al., 2008), is unique among reptiles because a number of species live in stable social groups comprising closely related individuals (Chapple,

2003). The only other lizard known to live in groups of related individuals is Xantusia vigilis (Davis et al, 2010). There is considerable variation in social organisation within the Egernia group, including solitary Liopholis inornata (Daniel, 1998), long-term monogamous pairs in Tiliqua rugosa (Bull, 2000), and!‘nuclear-families’!or extended family groups in Egernia saxatilis, Egernia cunninghami and Egernia stokesii

(Gardner et al., 2001; O’Connor and Shine, 2003; Stow and Sunnucks, 2004).

Intraspecific variation within and among populations of Liopholis whitii has also been described (Chapple and Keogh, 2005; While et al., 2009b). Such diversity makes the

Egernia group an ideal model for examining factors that may influence family living in lizards.

Constraints imposed by the availability of suitable habitat or resources in the environment may have led to the emergence of social living (Johnson et al., 2002;

Mouton, 2011). From here, it is likely that kin-based group living evolved because of inclusive benefits associated with sharing such resources with relatives (Hamilton,

1964). The presence of social aggregations in the Egernia group has been attributed to constraints such as this (O’Connor and Shine, 2003; Chapple and Keogh, 2005;

Michael et al., 2010). However, the presence of solitary species in similar habitat

! 90! ! ! ! ! ! ! (Langkilde and Shine, 2004) suggests that other (e.g. social) factors may also play an important role in maintaining group living.

All of the social Egernia group species studied to date rely on pre-existing shelter sites such as rock crevices or tree hollows (Stow et al., 2001; Duffield and

Bull, 2002; O’Connor and Shine, 2003; Duckett et al., 2012) or live in close association with them (Chapple and Keogh, 2006; While 2009b). As a result, populations are often highly saturated, and individuals aggressively compete for a limited number of retreat sites (Chapple, 2003; O’Connor and Shine, 2004), with high juvenile mortality and conspecific aggression (While et al., 2008a). Rather than being a response to dispersal constraints per se, delayed juvenile dispersal and the tolerance of offspring in the parental territory may be adaptive behaviours in themselves, because they result in prolonged association between offspring and parents, allowing juveniles access to resources for which they may otherwise be in intense competition (Covas and Griesser, 2007). This basic form of parental care – a precursor to kin-based sociality – is likely to be influenced by the mating system, because this affects genetic relatedness between adults and juveniles sharing a resource (While et al., 2009). An adult is more likely to tolerate, and care for, young with which it shares a higher level of genetic relatedness, because the inclusive fitness benefits of doing so increase as relatedness increases (Emlen, 1995; While et al., 2009a).

One species, Liopholis kintorei, may not be so constrained by retreat site availability because individuals excavate extensive burrow systems in which to live, although the availability of suitable habitat may still be a limiting factor. Extensive field observations of a population of L. kintorei at Uluru, Northern Territory, by

McAlpin (2001; 2011) and McAlpin et al. (2011) showed that individuals of varying sex and age classes may cohabit within a single burrow system. Genetic analysis of

! 91! ! ! ! ! ! ! this population showed that these ‘groups’ largely consisted of highly related individuals. Liopholis kintorei is a large, viviparous skink (187mm SVL; Wilson and

Swan, 2010), endemic to the central arid zone of Australia, nationally listed as

‘Vulnerable’!(IUCN, 2014). It largely inhabits sand plains and palaeodrainage lines and, while its range is extensive (approximately 1.3 million km2), it has a very patchy distribution, recorded at fewer than 100 localities within this area (McAlpin, 2011).

Liopholis kintorei is a nocturnal lizard and its elusive behaviour and underground lifestyle makes it difficult to observe in the wild.

The arid environment in which L. kintorei lives experiences extreme temperatures, low humidity, and unpredictable rainfall (Morton et al., 2011). In such an environment, food availability can be unpredictable, and so stable ‘fall back’!food resources are crucial for survival (James, 1991). Termite colonies represent concentrated food resources for arid-zone lizards (Shuttleworth et al., 2008), and although L. kintorei is an opportunistic feeder, the termite Drepanotermes perniger makes up a large proportion of its the diet, particularly when other food items are scarce (McAlpin, 2001). Drepanotermes perniger termites are endemic to Australia and unique among termites because they forage above ground, and rarely move more than a few metres from where they emerge out of galleries extending from their nests (Andersen and Jacklyn, 1993). Given the importance of these termites in the diet of L. kintorei, it is possible that such a resource would be an important factor in the maintenance of group living. Termite biomass in central Australian deserts is estimated to be higher than for any other herbivore (Watson et al., 1973), and without such an abundant food resource, competition between group members for food may become too great for groups to persist (Shuttleworth et al., 2008).

This study aimed to expand on McAlpin et al.’s (2011) work by:

! 92! ! ! ! ! ! ! (1) assessing the social organisation of great desert skinks at Newhaven (a site approximately 300 km north of the previously-studied Uluru population), which have been shown to be genetically distinct from the Uluru population (Dennison et al., 2015; Chapter 2 this thesis). McAlpin et al. (2011) assessed group stability by measuring relatedness among group members of varying age classes; we repeated this work at the Newhaven site, and expanded on it by utilising mark-recapture techniques over three consecutive years, in addition to genetic analysis, to investigate group membership, relatedness and stability; and

(2) establishing whether active burrows were predictable in space, based on features of the environment (i.e. termite presence) and therefore, if the environment may play a role in burrow placement.

Methods

Study site

Our 20 ha study site was located within the Australian Wildlife Conservancy’s

Newhaven Sanctuary, approximately 360 km northwest of Alice Springs, Northern

Territory, Australia (22° 48’!S, 131° 15’!E). In this area, L. kintorei burrows are located in sandy spinifex grassland, dominated by hummock grass (Triodia pungens) and the shrub, desert honey myrtle (Melaleuca glomerata). On average, temperatures range from approximately 23 ºC to 38 ºC in summer, and 7 ºC to 30 ºC in winter, although more extreme temperatures often occur (occasional frosts occur in winter, and we recorded a maximum shade temperature of 50 ºC in the austral summer of 2012).

Rainfall is highly variable, but most common in the summer months, with the area receiving approximately 362 mm annually (Australian Bureau of Meteorology, http://www.bom.gov.au).

! 93! ! ! ! ! ! ! Field methods

To assess group membership and stability, we carried out a mark-recapture study over three years (2011–2013) between October and November, during the active breeding period of L. kintorei. In 2012, sampling was most intensive and extended into late December. Thirty-one active burrow systems (discrete complexes of interconnected tunnels) were identified by the presence of L. kintorei tracks, and fresh scats in the external, above-ground latrine. Every burrow system at the site was

GPS-marked, checked for activity, and thereafter trapped. Baited Elliott traps were placed at every entrance of active burrows before dusk on each trapping night, and checked early the next morning (average 3064 trap-nights per season).

Upon initial capture we took body size measurements and tissue samples from each lizard. Morphometric measurements included head width and head length

(±0.1 mm) and snout-vent length (SVL; ±1 mm). The sex of all mature individuals

(>160 mm; Storr, 1968) was determined by hemipene eversion. Sex of immature individuals could not be determined. Tissue samples were collected for genetic analyses of parentage and group-relatedness. We obtained these by removing approximately 1 cm from the tail-tip, and preserving the tissue in 90% ethanol. We then inserted a subcutaneous passive integrated transponder (PIT-tag) for identification of individuals upon recapture. All lizards were returned to their point of capture after processing.

Liopholis kintorei is an elusive species and trapping success was not always high. To complement trapping, remote PIT-tag readers (LID650; Trovan Ltd) were deployed in 2012 and 2013. These consisted of circular aerials (10 cm diameter) positioned around the perimeter of the tunnel entrances to record the presence of any PIT-tagged lizards passing through. We positioned four readers around tunnel

! 94! ! ! ! ! ! ! entrances per burrow system, and rotated these around active burrow systems for three to four nights at a time, resulting in a total of 453 reading nights.

Assessing social organisation

Previous studies of social organisation in Egernia group species have determined animal associations in real time by either observing them within their rocky retreat sites (crevices; Duffield and Bull, 2002; Chapple and Keogh, 2006) or tracking their movements using GPS units attached to the lizard (Leu et al., 2010; Leu et al., 2011).

Both of these methods are difficult with L. kintorei, because we cannot directly observe animals within their burrows (due to their threatened status we could not disrupt the burrow systems), and GPS units interfere with the lizards’!burrow use (D.

Moore, pers. comm.). Therefore, our observations of social organisation were restricted to animals trapped at burrow systems overnight or recorded by remote PIT- tag readers. Animals that were observed ≥!3 times at a burrow system in a single year were said to be inhabitants of that burrow, and assumed to be part of that

‘group’. Individuals captured < 3 times were assumed to be transients and excluded from subsequent analyses. Animals caught < 3 times in a single year were included as group members only if they were captured from the same burrow ≥ 3 times in the previous or following year. The high proportion of recaptures, both within and across years (see Figures A2–4 in the Appendix), suggests that the majority of individuals at the site were captured and sampled. However, given the low observation rate of L. kintorei, analyses of within-group relatedness and composition were also repeated, using a less restrictive dataset that included all individuals observed as group members of the burrow system at which they were observed (and not just those captured ≥ 3 times; Supplementary material 4.2).

! 95! ! ! ! ! ! ! Molecular methods and summary statistics

We used a modified salting out protocol to extract DNA (Sunnucks and Hales, 1996), and genotyped all individuals using eight previously-developed microsatellite loci

(EST1, EST2, EST9 – Gardner et al., 1999; ECU3 – Stow, 2002; BX6, CKD, FQR,

J3F – Dennison et al., 2015), amplified using the protocol outlined in Dennison et al.

(2015; Chapter 2 this thesis).

Microsatellite alleles were visualised and scored using Peak Scanner 1.0

(Applied Biosystems). To ensure amplification and scoring consistency, at least 10% of samples were independently rerun and genotyped at each locus. Summary statistics including tests for Hardy-Weinberg equilibrium and linkage disequilibrium were carried out using GENALEX v6.5 (Peakall and Smouse, 2012), GENEPOP

(Raymond and Rousset, 1995) and FSTAT v2.9 (Goudet et al., 2002).

Parentage assignment

Assigning immatures to cohorts

Liopholis kintorei groups contain adults as well as immature individuals of discrete size classes. To determine the year of birth of immature individuals, and confidently exclude young individuals as putative parents, we calculated a logistic growth curve based on SVL measurements from our mark-recapture data (see Supplementary material 4.1 for methods). These data were also used to infer breeding characteristics (e.g. mate fidelity) of adults within and across seasons.

Parentage analyses

All individuals captured across the site were included in parentage analyses. We assigned parents to all immature individuals (N = 34) captured using the maximum- likelihood methods implemented in CERVUS 3.0 (Kalinowski et al., 2007) and

! 96! ! ! ! ! ! ! COLONY (Jones and Wang, 2010). Analyses were run separately for each year in which immature individuals were born. All adults captured at the site were included as potential parents for all immature individuals. Like other species within the Egernia group, L. kintorei is thought to be long-lived and exhibit high levels of philopatry

(Duffield and Bull, 2002; Stow and Sunnucks, 2004; McAlpin et al., 2011). We therefore assumed that parents of the juveniles born prior to the commencement of this study would still be present at the site, and attempted to assign parentage for these individuals as well. Lizards born in the same year or captured as immatures in the same year were excluded as potential parents, and those that matured during the study were only considered as potential parents for the youngest individuals (born in

2012). For the CERVUS analysis, we ran 10 000 cycles, with proportions of 0.8 candidate parents sampled, 0.98 loci typed and a genotyping error rate of 0.01. The

COLONY analysis was run conservatively allowing for both sexes to be polygamous, no sibship prior, and 0.8 probability that candidate parents had been sampled.

Parentage was accepted if: the parent–offspring dyad had mismatches at no more than one locus; and the confidence of the match was 0.95 (strict) or 0.80

(relaxed) in CERVUS, or this result was consistent with the COLONY analysis.

Relatedness estimations

Relatedness between individuals was estimated in the program COANCESTRY v1.0

(Wang, 2011). We calculated the pairwise Queller and Goodnight (1989) index of relatedness (R) between all individuals. We then tested for differences in within- and among-group relatedness for all sex and age classes, using the bootstrapping method implemented in COANCESTRY. To indirectly assess group stability, we also examined pairwise relatedness of group members pooled across years using the

Pops as Dclass approach implemented in Genalex 6.5 (Peakall and Smouse 2012).

! 97! ! ! ! ! ! ! If R between group members across years was not significantly different from zero, then we could conclude that group membership was changing across years.

Analyses were carried out first on the more conservative dataset of those individuals captured ≥ 3 times at a burrow system, and then repeated including all animals captured at the site (Supplementary material 4.2).

To assess whether females preferentially mated with unrelated males, we compared relatedness between a focal female and her mate (as identified by parentage analysis) with her average pairwise relatedness to all other males captured at the site. A randomised paired t-test was performed with 5000 permutations to evaluate the difference between these values.

Assessing attributes of burrows and their environmental correlates

In October–November 2012, we carried out surveys at 21 active burrow systems. We mapped all active tunnels and latrines, and measured the length (m) between the most distant tunnel entrances. We then placed a 4 m ×!4 m quadrat over the burrow system in which we visually estimated the percentage of ground cover (separated into % ground cover < 1 m, and shrubs > 1 m in height). To determine whether burrow systems were associated with the presence of termite mounds (a stable food resource), we also measured the distance from the centre of each quadrat to the closest termite mound. These data were compared with 20 control quadrats (i.e. burrow absent) placed at the site by randomly selecting UTM coordinates within the range of the site, and finding them by GPS in the field. MANOVA was used to determine significant differences between burrow and control sites, with percentage of ground cover, shrub cover, and proximity to termites (distance in m) as dependent variables, and presence/absence of a burrow system as the independent variable.

We also tested whether these environmental correlates differed between burrows

! 98! ! ! ! ! ! ! that were identified as containing groups in 2012, and those from which only one individual was captured (group burrows vs. solitary burrows).

Results

Parentage assignment

A total of 65 individual lizards were captured over the three years of sampling (Table

4.1). Of these, 34 were initially captured as immatures, and we attempted to assign parentage to these.

Table 4.1. Number of Liopholis kintorei individuals of each age and sex class captured in each year of sampling. The annual totals include recaptures from previous years. The overall total represents the number of individual lizards captured by the end of the study

Year! Adult females! Adult males! Immatures! Total!

2011! 6! 8! 25! 39!

2012! 26! 15! 7! 48!

2013! 15! 7! 9! 31!

No. unique individuals! 28! 16! 21†! 65!

† 21 individuals were still classified as juveniles by the end of the study; 34 animals were initially captured as juveniles (as in the text), however some of these matured during the course of the study, and so they are included in the adult counts in the bottom row of the table depicting unique individuals.

None of the loci showed significant linkage disequilibrium or deviation from Hardy-

Weinberg equilibrium (see Table 3.3, Chapter 3 this thesis). From the juveniles captured, litter sizes of 1–4 were identified based on genetic identification of siblings, however Pearson et al. (2001) reported litter sizes of 1–7 in L. kintorei from captive- held animals. Parent pairs were assigned to 20 out of 34 (58.8%) of the offspring sampled (seven parent pairs). A further six offspring (17.6%) were assigned to mothers only. Three mothers were assigned to > 1 cohort of offspring, and all had

! 99! ! ! ! ! ! ! mated with the same male across more than one breeding season. One of these mothers changed mates in the final year of sampling (third cohort of offspring), although her original mate was not recaptured and may have not survived into this final breeding year. No males were found to have mated with more than one female, and only one female appeared to have mated with more than one male in a single season. This was determined by the presence of maternal half-siblings from a single age cohort, as the second mated male was not sampled.

There was no evidence of disassortative mating based on relatedness (t =

0.807, P = 0.577). Relatedness was not significantly different between females and their mates (R = -0.0712, SE = 0.0946) and females and non-mates (R = -0.0144, SE

= 0.0197), however this may be due to the small sample size (N = 7 mated pairs).

Group composition and stability

Using the conservative dataset (individuals captured ≥ 3 times at a burrow) we identified a total of 11 social groups throughout the study (Table 4.2), ranging in size from 2 to 9 individuals (overall mean group size ± SE = 3.6 ± 0.2). When all individuals at the site were included in analyses, mean group size did not change substantially (Supplementary material 4.2). Those considered ‘residents’!of the study site were captured an average of 14.5 (SE = 1.2) times during the study. We assigned 23 individuals (59%; 3 females, 2 males, 18 immatures) to groups in 2011,

40 (83%; 20 females, 12 males, 8 immatures) in 2012, and 22 (71%; 9 females, 4 males, 9 immatures) in 2013.

Group composition varied and membership was dynamic, although most groups had core individuals that remained in the same group across consecutive years. Two groups (B19, 23) consisted of single adult male–female pairs present only in 2012. B17 contained related female adults (R = 0.55; equivalent of mother–

! 100! ! ! ! ! ! ! daughter or full-sibs) in 2012, but only one of those females and her offspring in

2013. B45 consisted of an adult female and her two offspring in 2012 and 2013. B13 consisted of an adult pair with their offspring in 2011 and 2012, but in 2013 the adult pair was not recaptured. Two of the immatures remained in the group with a new adult pair. B24 comprised a single adult female and four immatures in 2011, two of which matured in the burrow and were present until the end of the study. In 2012 an adult male was present in this group, as well as a new juvenile that was also present the following year. B2 comprised three immatures in 2011, one of which matured and was the sole inhabitant in 2012. In 2013, a new adult pair had taken over the burrow.

B36 comprised an adult female and two males, although the female and one of the males appeared to move around together, captured from two adjacent burrow systems (including one occupied by another group) without the second male. B14 included an adult male and a juvenile in 2011, two adult males in 2012, and neither was recaptured in 2013. B1 contained four adult females (two of which were present the previous year, one a subadult), a juvenile and an adult male. The male and one of the females were the parents of the subadult, and the female had two other adult offspring within the group. Finally, B5 contained only family members: a single female was identified as the mother of all other group members except the adult male

(present in 2012), who was father to her third cohort of offspring (present in 2013).

Throughout the study, 11 individuals were associated with this group, and all but two were present across two or more years.

Some individuals were observed as solitary (six in 2011, six in 2012, and five in 2013). Of these, two solitary adult females were captured in the same burrow in all three years of sampling, one female was captured in only two years, but in the same burrow, and one male was captured in several different burrows within and across years. Given the low observation rate of this species and variability in sampling

! 101! ! ! ! ! ! ! success, it is possible that individuals categorised as solitary may have been cohabiting with other individuals who were not captured. However, given that the same individuals were categorised as solitary in more than one year, we assumed that this was their true social status. Within seasons, groups were stable, with only two individuals (4.2%) in 2012 being observed > 3 times in a group other than the one they were assigned to.

Throughout the study, 32.4% of immature individuals assigned to parents were captured in the same burrow as both parents in the same year, 23.5% were captured in the same burrow as their mother only, 2.9% with the father only, and

73.5% were captured from the same burrow as at least one of their full siblings.

Relatedness within and among groups

Average pairwise R between group members (within-groups) was significantly higher than that between non-group members (among-groups) in all three years tested

(Table 4.3; average within-group relatedness across all years ± SE = 0.292 ± 0.01), and this was also the case when all individuals were included in analyses

(Supplementary material 4.2). B5 was a large group comprised of full and half siblings. To ensure that inclusion of this group was not driving up average within- group R, we also ran analyses excluding B5. Within-group R was still significantly higher than among-group R, and so results reported here include B5. When only adults were considered, within-group R was significantly higher than among-group R in 2012 and 2013, but not in 2011, but this is likely due to the small number of pairwise comparisons in that year (Table 4.3). This was also the case for adult females, and there was insufficient data in all years to draw conclusions on the within- and among-group relatedness for adult males. When relatedness was

! 102! ! ! ! ! ! ! compared between all group members pooled across years, 3 out of the 9 groups had significantly higher R than that expected by chance (Figure 4.1).

Overall, the results did not change substantially when relatedness calculations included all animals sampled (see Supplementary material 4.2 for detailed explanation).

! 103! ! ! ! ! ! ! Table 4.2. Composition of Liopholis kintorei groups at Newhaven in 2011–2013. The number of individuals belonging to each sex and age class are given for each year of study, as well as the total number of animals captured and average pairwise relatedness for each group

Group! Year! Adult females! Adult males! Immatures! Total! Within-group R (SE)!

B1! 2011! 1! 0! 1! 2! 0.0737 (0.00)! ! 2012! 4! 1! 1! 6! 0.1275 (0.08)!

! 2013! -! -! -! -! -! B2! 2011! 0! 0! 3! 3! 0.1151 (0.15)! ! 2012! -! -! -! -! -! ! 2013! 1! 1! 0! 2! -0.2304 (0.00)!

B5! 2011! 0! 0! 6*! 6! 0.5306 (0.03)! ! 2012! 5! 1! 3*! 9! 0.5254 (0.03)!

! 2013! 2! 0! 3*! 5! 0.5971 (0.02)!

B13! 2011! 1! 1! 3*! 5! 0.2515 (0.09)!

! 2012! 2! 1! 1! 4! -0.0764 (0.12)!

! 2013! 1! 1! 2*! 4! -0.0569 (0.06)!

B14! 2011! 0! 1! 1! 2! 0.4039 (0.00)!

! 2012! 0! 2! 0! 2! -0.0408 (0.00)!

! 2013! -! -! -! -! -!

B17! 2011! -! -! -! -! -!

! 2012! 2! 0! 0! 2! 0.5488 (0.00)!

! 2013! 1! 0! 1! 2! 0.5409 (0.00)!

B19! 2011! -! -! -! -! -!

! 2012! 1! 1! 0! 2! -0.0993 (0.00)!

! 2013! -! -! -! -! -!

B23! 2011! -! -! -! -! -! ! 2012! 1! 1! 0! 2! -0.4161 (0.00)! ! 2013! -! -! -! -! -!

B24! 2011! 1! 0! 4! 5! 0.0982 (0.07)!

! 2012! 3! 1! 1! 5! 0.0151 (0.05)! ! 2013! 2! 0! 1! 3! 0.0265 (0.03)!

B36! 2011! -! -! -! -! -!

! 2012! 1! 2! 0! 3! 0.1127 (0.09)! ! 2013! 1! 2! 0! 3! 0.1127 (0.09)!

! 104! ! ! ! ! ! !

Group! Year! Adult females! Adult males! Immatures! Total! Within-group R (SE)!

B45! 2011! -! -! -! -! -! ! 2012! 1! 0! 2! 3! 0.4233 (0.02)! ! 2013 1! 0! 2! 3! 0.4233 (0.02)!

Solitary ! 2011! 3! 1! 2! 6! -! ! 2012! 4! 2! 0! 5! -! ! 2013! 3! 1! 1! 5! -!

* denotes >1 age cohort of immatures within a group

! 105! ! ! ! ! ! ! Table 4.3. Average relatedness (R) between Liopholis kintorei within and among social groups at

Newhaven each year. Standard errors (SE) and P-values were calculated via bootstrapping. In all statistically significant cases, within-group relatedness was higher than among-group relatedness. “na”! indicates insufficient data for comparisons

! ! Within-groups! Among-groups!

! Year R! SE! R! SE! P-value!

All adults & immatures! 2011! 0.3070! 0.0444! -0.0285! 0.0119! <0.01! ! 2012! 0.2873! 0.0373! -0.0465! 0.0073! <0.01! ! 2013 0.2825! 0.0596! -0.0558! 0.0118! <0.01!

Adults only! 2011! na! na! -0.4491! 0.0405! na!

! 2012! 0.1836! 0.0519! -0.0299! 0.0095! <0.01! ! 2013 0.1172! 0.0894! 0.0312! 0.0216! <0.05!

Adult females! 2011! na! na! 0.1066! 0.0944! na!

! 2012! 0.3315! 0.0641! -0.0421! 0.0138! <0.01!

! 2013 0.3306! 0.1683! -0.0654! 0.0242! <0.01!

Adult males! 2011! na! na! na! na! na!

! 2012! -0.0144! 0.0187! -0.0037! 0.0380! NS!

! 2013 na! na! 0.21272! 0.1035! na!

Immatures! 2011! 0.3495! 0.0539! -0.0402! 0.0173! <0.01!

! 2012! 0.5991! 0.0637! -0.0717! 0.0278! <0.01!

! 2013! 0.5222! 0.0294! 0.0398! 0.0272! <0.01!

! 106! ! ! ! ! ! !

Fig. 4.1. Average pairwise relatedness between group members pooled across years. Red marks represent the 95% confidence interval around which relatedness is effectively zero

Burrow presence, but not group presence is predicted by environmental attributes

On average, burrow complexes spanned an average of 8.12 m (SE = 0.39, range =

5.6–11.7) at their widest point, had 7.10 tunnel entrances (SE = 0.84, range = 3–17) and 1.19 latrines (SE = 0.09, range = 1–2). Burrow systems were significantly closer to termite mounds, and had significantly more ground cover but fewer shrubs than control quadrats (i.e. no burrow present; F3,37 = 12.97, P < 0.0005, Wilk’s Λ = 0.487,

η2 = 0.51; Figure 4.2; Table 4.4). There was no significant difference in any of these environmental attributes when comparing burrows occupied by groups with those

2 occupied by solitary individuals (F1,19 = 1.65, P = 0.216, Wilk’s Λ = 0.775, η = 0.23;

Table 4.5).

! 107! ! ! ! ! ! !

Distance)to) %)ground) %)shrub) termite)mound) cover) cover) Burrow"absent" Burrow"present"

Fig. 4.2 Heat map illustrating the differences in distance to termite mounds, % ground cover and % shrub cover for burrow and control sites. Each row represents one quadrat. The darker the colour of the cell, the greater the value for that variable

! 108! ! ! ! ! ! ! Table 4.4. Summary statistics (MANOVA) of the distance to the nearest termite mound from each burrow complex or control quadrat, and ground cover within the quadrat (% vegetation cover < 1 m, and % cover of shrubs > 1 m)

Burrow Control F df P Partial Variable! N! (mean ±"SE)! N! (mean ±"SE)! η2

Distance (m)! 21! 3.68 ±!0.38! 20! 5.41 ±!0.68! 5.09 1,39 0.03 0.12

Ground cover (%)! 21! 36.19 ±!3.20! 20! 18.25 ±!2.03! 21.95 1,39 0.00 0.36

Shrubs (%)! 21! 2.86 ±!2.09! 20! 25.25 ±!5.30! 16.02 1,39 0.00 0.29

Table 4.5. Summary statistics (MANOVA) for burrows containing groups vs. solitary individuals in

2012. Distance to the nearest termite mound from each burrow complex is given, as well as ground cover within the quadrat (% vegetation cover < 1 m, and % cover of shrubs > 1 m)

Group Solitary burrows burrows F df P Partial Variable! N! (mean ±"SE)! N! (mean ±"SE)! η2

Distance (m)! 16! 3.54!±!0.42! 5! 4.10!±!0.92! 0.39 1,19 0.54 0.02

Ground cover (%)! 16! 33.75!±!3.01! 5! 44.00!±!9.28! 1.95 1,19 0.18 0.09

Shrubs (%)! 16! 1.25!±!1.25! 5! 8.00!±!8.00! 1.99 1,19 0.17 0.10

Discussion

This study shows high variation in social organisation of L. kintorei, with group composition ranging from single adult pairs to family groups containing more than one age cohort of offspring. Group composition was stable within seasons, but appears to be variable across years. It is possible that incomplete sampling may have lead us to overestimate this variability because sampling success varied across years. The fact that animals were not captured from every burrow each year might be a reflection of the difficulties of sampling this elusive species, and we cannot be

! 109! ! ! ! ! ! ! certain that non-capture truly meant that a burrow was vacant. However, the fact that average pairwise relatedness remained higher than expected by chance in 33% of burrows when group members were pooled across years (Figure 4.1) suggests that some groups are in fact more temporally stable than others, or that L. kintorei preferentially associate with kin. Although social polygyny was evident, no males were found to have mated multiply with females – a result in stark contrast with

McAlpin et al.’s (2011) finding of 40% polygynous males at Uluru. This result was surprising, given that males at this location have been previously found to move among burrows more than females, presumably looking for extra-pair mating opportunities (Chapter 3 this thesis). The level of within-group relatedness in this study was comparable, though slightly lower than that described at Uluru as well

(McAlpin et al., 2011), which is also unexpected if genetic polygyny is in fact considerably lower at Newhaven. However, only 7 mated pairs could be confirmed by parentage analysis in this study, which may not be representative of this population.

Relatedness was, on average, higher within groups than among groups in all years. Females and juveniles were more highly related within than among burrows, although this was not the case for males using either the conservative or less- restricted datasets. Liopholis kintorei burrows were positioned in areas with relatively higher percentage ground cover (< 1 m) and lower shrub cover (> 1 m) than control sites, and were on average in closer proximity to termite mounds, an important food resource for this species.

Liopholis kintorei lives in small groups (2–9) comparable to those found in

Liopholis whitii (2–6; Chapple and Keogh, 2006), and Egernia saxatilis (2–6;

O’Connor and Shine, 2003). Group membership was dynamic across years in many cases, however 6 out of the 11 (55%) groups identified had at least two members present across two or more years, and one of the groups (10%) had at least two

! 110! ! ! ! ! ! ! members present in all three years of sampling. Although the majority of lizards captured could be assigned to groups, there was high variability in capture rates of individuals (1–20 times in a single year), and trapping success was highest in 2012.

This may have led to an underestimation of group membership and stability.

However, the fact that within-group relatedness remained significantly high when individuals were pooled across years (Figure 4.1), suggests membership is relatively stable for some groups.

The presence of adult daughters in several of the burrows also suggests longer-term stability of these groups. For example, in 2011, the group B5 contained immatures estimated from our growth curve (Supplementary material 4.1) to have been born in 2008. These were still present as adult females in 2012, suggesting that they had remained in the natal burrow for at least four years, the mother likely longer.

In 2011, group B13 contained a parent pair with two offspring born in 2010. The same parent pair and one of the offspring were still together in the same burrow in

2012, a minimum of two years since birth. This agrees with McAlpin et al. (2011), who estimated that on average, burrow occupancy spans approximately four age cohorts in this annually breeding lizard.

In all years, average pairwise relatedness within groups was significantly higher than that among groups (Table 4.4), although average within-group relatedness was variable (Table 4.3). Average within-group relatedness values were consistent with those recorded by McAlpin et al. (2011) at Uluru (average within-group R at Uluru =

0.371). Adult females were more closely related to other within-group females than those from other groups, as were immatures. Using the conservative dataset, data was insufficient to assess this in males in 2011 and 2013, although in 2012 there was no significant difference in the relatedness of males within and among groups (Table

4.4), however the less-restricted dataset also corroborated this (Supplementary

! 111! ! ! ! ! ! ! material 4.2). This is consistent with the finding of male-biased dispersal and female philopatry in this species (Chapter 3 this thesis). The high within-group relatedness of females and presence of adult daughters in burrows evident in this study suggest that they (and their offspring) might gain inclusive fitness benefits from cohabiting with relatives. Such benefits identified in other Egernia species include access to potentially heritable, high quality resources (Langkilde et al., 2007), protection from predators through enhanced vigilance within the group (Lanham and Bull, 2004) or sharing of communal retreat sites (McAlpin et al., 2011), and reduced conflict from conspecific males (While et al., 2009a). Additionally, philopatry may enable female

L. kintorei to invest less in searching or competing for refuges, and more in searching for food and in reproduction, while males may move among burrows more, and traverse larger distances to secure matings with multiple females. This is supported by the very low within-group relatedness found among males. As well as increasing survival and thus inclusive fitness within the group, delayed dispersal of adult females may serve to reinforce mother–offspring or sibling bonds, a precursor to kin- based social living (Davis et al., 2011).

Previous studies in other social species of the Egernia group have shown high stability of groupings across years (Duffield and Bull, 2002; O’Connor and Shine,

2003; Chapple and Keogh, 2006). One characteristic of these species is that they are generally associated with pre-existing permanent retreat sites (Chapple, 2003;

Michael, 2010). The limited availability of high quality retreat sites has been suggested as an important factor in the emergence of kin-based social aggregation in

Egernia (Lanham, 2001; Michael et al., 2010). Often these refuges are patchily distributed in the environment, separated by expanses of unsuitable habitat (Duffield and Bull, 2002; D. O’Connor and Shine, 2003). Because L. kintorei is an obligate burrowing species, it could be assumed that the apparent homogeneity of the

! 112! ! ! ! ! ! ! spinifex-dominated sandplains in which they live could mean that individuals are unlikely to be so constrained by the availability of retreat sites, since they can construct their own (McAlpin et al., 2011). However, groups were generally stable within seasons, although less so across years, and groups still contained relatives.

This could reflect two things: (1) despite being able to construct their own retreat sites, L. kintorei are still constrained by the availability of suitable habitat in which to do so, (as suggested by our environmental analyses); and/or (2) factors other than habitat availability, such as social benefits, may also contribute to the maintenance of kin-based group living (discussed above).

It is likely that these groupings are a result of juveniles delaying their dispersal. In so doing, they obtain access to, and may even inherit the natal burrow and its associated resources. This would have direct survival benefits, given the high level of protection afforded by burrows in buffering inhabitants from extreme temperatures and in protection from predators (Chapple, 2003; Fenner et al., 2012). The apparent reduced pressure for adults to find an existing shelter site, when one may be constructed, could contribute to the reduced stability of some groups across years, compared with other Egernia group species. However, burrow construction in itself represents a significant investment of time and energy, and simply building a new shelter site when required may not be so easy, nor might it be as high-quality a construction as those burrows that have been established over multiple generations.

Indeed, McAlpin (2001) states that most newly established L. kintorei burrows fail, probably because small, simple burrows may offer less protection than the large complex tunnel systems characteristic of the older, more established burrow complexes. Additionally, other important resources, such as a stable food source, may bind L. kintorei to a particular site, and may facilitate or maintain kin-based social living.

! 113! ! ! ! ! ! ! According to the Ecological Constraints Hypothesis, social grouping can be driven by limited resource availability (Emlen, 1982). The extreme and unpredictable desert environment that L. kintorei inhabits highlights the importance of a stable and reliable food resource for survival in this region (Morton et al., 2011). We found that active burrows were significantly associated with termite mounds, suggesting that these resources may be important in the persistence of L. kintorei groups. Indeed, the dependence on termites as a food source, particularly during dry periods, has been proposed as a direct cause of group living in the desert-dwelling cordylid lizard,

Ouroborus cataphractus (Shuttleworth et al., 2008; Mouton, 2011). Without a concentrated food source such as Drepanotermes nests, it is possible that intra- group competition for food would be too high for L. kintorei to maintain group living.

While broader-scale surveys across multiple sites where L. kintorei are present and absent would be required to assess the relationship between termite density and group size and stability, this was beyond the scope of our work. However, our finding that burrows are closely associated with termite mounds at our site is biologically relevant. When termite alates swarm (winged, reproductive males and females emerging from the nest during their nuptial flight), there is a rapid convergence of animals to the termite colony to opportunistically feed. These events are unpredictable and animals must respond immediately to feed when they occur (Dial and Vaughan, 1987). Liopholis kintorei would benefit from being within close range of a termite mound to take advantage of an emergence event when it occurs, and could rapidly retreat to their burrows if a predator was attracted by the concentrated foraging of numerous animals during these times (Dial and Vaughan, 1987).

It is unclear how biologically significant is a difference of 2 m in distance to termite nests (3.6 m from a burrow vs. 5.4 m from a control quadrat, this study) in terms of fitness for L. kintorei. However, distance from a refuge is an important risk factor

! 114! ! ! ! ! ! ! when foraging, because individuals are more likely to be captured before reaching cover if they are further away to begin with (Cooper and Whiting, 2007). Castilla and

Labra (1998) showed a substantial increase in experimental predation as proximity to a predator decreased, with an approximately 30% increase in predation when lizard models were placed just 2 m closer (8 m vs. 6 m) to a potential predator. If individual

L. kintorei can detect a predator in close proximity before it is detected itself, this may be a sufficient difference to ensure successful flight to the burrow. Additionally, the amount of cover in the environment can have a large impact on the ability of predators to detect prey before attacking, and the ability of individuals to hide once an attack has been initiated (Cooper, 2003). For example, desert iguanas are harder to detect in bushy microhabitats than in areas with more sparse vegetation (Cooper,

2003). The higher percentage of ground cover surrounding L. kintorei burrows compared with control sites may provide protection from predators during activity around the burrows.

Competition among kin should drive dispersal from the natal territory unless the associated costs are balanced by the direct and inclusive benefits of remaining with relatives (Perrin and Mazalov, 2000; Davis Rabosky et al., 2012). Group cohesion may be facilitated by social interactions among conspecifics. For example,

Liopholis whitii males only tolerate their own genetic offspring in their territories, and the associated increased risk of infanticide may prevent females from seeking out mates outside the social group (While et al., 2009a). The construction and maintenance of the burrow systems in which L. kintorei live represent a significant energy investment (McAlpin et al., 2011), and it is possible that residents may exclude unrelated individuals, thus increasing inclusive fitness by sharing their burrow with kin (Emlen, 1995).

Many of the advantages of group living have been demonstrated in lizards,

! 115! ! ! ! ! ! ! such as increased vigilance (Lanham and Bull, 2004), indirect protection of juveniles

(O’Connor and Shine, 2004; While et al., 2009), heat retention and homeostatic benefits (Davis Rabosky et al., 2012). All of these are potential benefits to group living in L. kintorei. In particular, delayed juvenile dispersal and thus access to the natal burrow provides protection from environmental extremes in the desert environment (Fenner et al., 2012), and predators (McAlpin et al., 2011), as well as access to local food resources within the natal territory. The presence of related adult females may also have more direct benefits, as Sinn et al. (2008) showed that offspring survival was directly related to the extent of female aggression.

The presence of related adults and delayed juvenile dispersal in L. kintorei is a pattern concordant with those described in other species within the Egernia group

(Gardner et al., 2001; Stow and Sunnucks, 2004; Chapple and Keogh, 2005; While et al., 2009b). This suggests that the tendency for kin-based group living might be phylogenetically conserved within the lineage (Gardner et al., 2008). However, the evident variation in social organisation suggests a high level of individual variation in social strategies. Factors that might maintain this individual variation could include habitat quality and availability, as well as the availability of high-quality food resources. Although burrows were consistently found in close proximity to termite mounds, this association with a stable food source alone is not sufficient to explain social living in L. kintorei. Many desert lizards feed on termites (Morton and James

1988; James, 1991), and yet do not exhibit kin-based group living. For example, the sympatric and closely related L. inornata is a solitary burrowing species, which (given the likely plesiomorphy of sociality in Egernia) appears to have lost its sociality, while the trait has been maintained in L. kintorei. This could be due to less constrained habitat requirements of L. inornata, which constructs small burrows in sandy or loamy soils, however is found in a range of habitat types, with a much broader distribution

! 116! ! ! ! ! ! ! spanning both arid and more temperate climates (Daniel, 1998). In the case of L. kintorei, the presence of a reliable food source may, however, reduce the pressures of competition among kin in the unpredictable desert environment, and allow group living to be maintained.

Our study shows that L. kintorei live in groups of varied composition, although groups of related individuals that were relatively stable within seasons were evident.

Stability across years is variable, ranging from a single season to at least four years at our site. The variation in social organisation within and among species of the

Egernia group of lizards (Bull, 2000; Gardner et al., 2001; O’Connor and Shine, 2003;

Stow and Sunnucks, 2004; Chapple and Keogh, 2006), as well as within-populations

(While et al., 2009b), provides valuable opportunities to investigate factors promoting the evolution of group living. Liopholis kintorei in particular warrants further investigation, given the variation in social organisation demonstrated in this study, and its use of self-constructed burrows rather than pre-existing shelters. It is extremely difficult to thoroughly investigate associations among group members due to their elusive, underground lifestyle and threatened status. However, long-term assessment of group organisation, structure and stability throughout the year would shed light on the relative importance of food and shelter-site availability, and the extent to which these influence kin-based group cohesion.

Examining which factors may influence individual variation in social strategies within species will increase our understanding of, and provide insight into, the causes of diversity in social behaviours among species.

! 117! ! ! ! ! ! ! Acknowledgements

We would like to thank everyone who helped with sample collection: M. Asmyhr, A.

Beattie, H. Baldwin, J. Bishton, J. Davis, J. Dennison, M. Gillings, W. Greene, J.

Porter, P. Momigliano and C. Turnbull. In particular, thankyou to S. McAlpin for sharing his knowledge of great desert skinks, and to D. Moore and J. Schofield from the Australian Wildlife Conservancy for immense logistical support and local knowledge on site. All animals were handled in accordance with recommendations of

Macquarie and Charles Darwin universities’ animal ethics committees (ARA

2011/037), and sample collection was licensed by the Northern Territory Government

(NRETAS permit number 41144). This project was funded by Macquarie University, the Joyce W. Vickery Fund (to SD), the Rice Memorial Fund (to SD), and the

Australian Wildlife Conservancy.

Literature cited

Alexander, R. D. 1974. The evolution of social behavior. Annual Review of Ecology

and Systematics 5:325–383.

Altizer, S., Nunn, C., Thrall, P., Gittleman, J., Antonovics, J., Cunningham, A.,

Dobson, A., Ezenwa, V., Jones, K., Pedersen, A., Poss, M., and J. Pulliam.

2003. Social organisation and parasite risk in mammals: integrating theory and

emprical studies. Annual Review of Ecology, Evolution, and Systematics

34:517–547.

Andersen, A., and P. Jacklyn. 1993. Termites of the Top End. CSIRO Publishing,

Australia.

! 118! ! ! ! ! ! ! Blundell, G. M., Ben-David, M., and R. T. Bowyer. 2002. Sociality in river otters:

cooperative foraging or reproductive strategies? Behavioural Ecology 13:134–

141.

Bull, C. M. 2000. Monogamy in lizards. Behavioural Processes 51:7–20.

Chapple, D. G. 2003. Ecology, life-history, and behavior in the Australian Scincid

genus Egernia, with comments on the evolution of complex sociality in lizards.

Herpetological Monographs 17:145–180.

Chapple, D. G., and J. S. Keogh. 2005. Complex mating system and dispersal

patterns in a social lizard, Egernia whitii. Molecular Ecology 14:1215–1227.

Chapple, D. G., and J. S. Keogh. 2006. Group structure and stability in social

aggregations of white’s skink, Egernia whitii. Ethology, 112:247–257.

Clark, R. W., Brown, W. S., Stechert, R., and H. W. Greene. 2012. Cryptic sociality in

rattlesnakes (Crotalus horridus) detected by kinship analysis. Biology Letters

8:523–525.

Covas, R., and M. Griesser. 2007. Life history and the evolution of family living in

birds. Proceedings of the Royal Society B: Biological Sciences 274:1349–1357.

Daniel, M. C. 1998. Aspects of the ecology of Rosen’s desert skink, Egernia inornata,

in the Middleback Ranges, Eyre Peninsula. Department of Environmental

Science and Management. University of Adelaide, Roseworthy Campus.

Davis, A., Corl, A., Surget-Groba, Y., and B. Sinervo. 2010. Convergent evolution of

kin-based sociality in a lizard. Proceedings of the Royal Society B: Biological

Sciences 278:1507–1514.

Davis Rabosky, A., Corl, A., Liwanag, H., Surget-Groba, Y., and B. Sinervo. 2012.

Direct fitness correlates and thermal consequences of facultative aggregation in

a desert lizard. PLOS ONE 7:e40866.

! 119! ! ! ! ! ! ! Dial, K., and T. Vaughan. 1987. Opportunistic predation on alate termites in Kenya.

Biotropica 19.185–187.

Dickinson, J., and A. McGowan. 2005. Winter resource wealth drives delayed

dispersal and family-group living in western bluebirds. Proceedings of the

Royal Society B: Biological Sciences 272:2423–2428.

Doody, J. S., Burghardt, G. M., and V. Dinets. 2013. Breaking the social - non-social

dichotomy$: a role for reptiles in vertebrate social behavior research? Ethology

119:1–9.

Duckett, P. E., Morgan, M. H., and A. J. Stow. 2012. Tree-dwelling populations of the

skink Egernia striolata aggregate in groups of close kin. Copeia 2012:130–134.

Duffield, G. A., and C. M. Bull. 2002. Stable social aggregations in an Australian

lizard, Egernia stokesii. Naturwissenschaften 89:424–427.

Emlen, S. T. 1982. The evolution of helping. I. An ecological constraints model. The

American Naturalist 119:29–39.

Emlen, S. T. 1995. An evolutionary theory of the family. Proceedings of the National

Academy of Sciences USA 92:8092–8099.

Fenner, A. L., and C. M. Bull. 2010. The use of scats as social signals in a solitary,

endangered scincid lizard, Tiliqua adelaidensis. Wildlife Research 37:582.

Fenner, A. L., Pavey, C. R., and C. M. Bull. 2012. Characteristics of the burrows of

Slater’s skink, Liopholis slateri. Herpetological Journal 22:115–121.

Gardner, M. G., Cooper, S. J. B., Bull, C. M., and W. N. Grant. 1999. Isolation of

microsatellite loci from a social lizard, Egernia stokesii, using a modified

enrichment procedure. Journal of Heredity 90:301–304.

Gardner, M. G., Bull, C. M., Cooper, S. J., and G. A. Duffield. 2001. Genetic evidence

for a family structure in stable social aggregations of the Australian lizard

Egernia stokesii. Molecular Ecology 10:175–183.

! 120! ! ! ! ! ! ! Gardner, M. G., Hugall, A. F., Donnellan, S. C., Hutchinson, M. N., and R. Foster.

2008. Molecular systematics of social skinks: phylogeny and taxonomy of the

Egernia group (Reptilia: Scincidae). Zoological Journal of the Linnean Society

154:781–794.

Goudet, J., Perrin, N., and P. Waser. 2002. Tests for sex-biased dispersal using bi-

parentally inherited genetic markers. Molecular Ecology 11:1103–14.

Graves, B. M., and D. Duvall. 1995. Aggregation of squamate reptiles associated

with gestation, oviposition, and parturition. Herpetological Monographs 9:102–

119.

Griffin, A. S., and S. A. West. 2002. Kin selection: fact and fiction. Trends in Ecology

and Evolution 17:15–21.

Hamilton, W. D. 1964. The genetical evolution of social behaviour I. Journal of

Theoretical Biology 7:1–16.

Hughes, C. 1998. Integrating molecular techniques with field methods in studies of

social behavior: a revolution results. Ecology 79:383–399.

IUCN. (2014). IUCN Red List of Threatened Species. Version 2014.2. Retrieved from

www.iucnredlist.org.

James, C. D. 1991. Temporal variation in diets and trophic partitioning by coexisting

lizards (Ctenotus: Scincidae ) in central Australia. Oecologia 85:553–561.

Jarvis, J. U. 1981. Eusociality in a mammal: cooperative breeding in naked mole-rat

colonies. Science 212:571–573.

Johnson, D. D., Kays, R., Blackwell, P., and D. Macdonald. 2002. Does the resource

dispersion hypothesis explain group living? Trends in Ecology and Evolution

17:563–570.

! 121! ! ! ! ! ! ! Jones, O. R., and J. Wang. 2010. COLONY: A program for parentage and sibship

inference from multilocus genotype data. Molecular Ecology Resources 10:551–

555.

Kalinowski, S. T., Taper, M. L., and T. C. Marshall. 2007. Revising how the computer

program CERVUS accommodates genotyping error increases success in

paternity assignment. Molecular Ecology 16:1099–1106.

Langkilde, T., and R. Shine. 2004. Competing for crevices: interspecific conflict

influences retreat-site selection in montane lizards. Oecologia 140:684–91.

Lanham, E. J. 2001. Group-living in the Australian skink, Egernia stokesii. Flinders

University, South Australia.

Lanham, E. J., and C. M. Bull. 2004. Enhanced vigilance in groups in Egernia

stokesii, a lizard with stable social aggregations. Journal of Zoology 263:95–99.

Leu, S. T., Bashford, J., Kappeler, P. M., and C. M. Bull. 2010. Association networks

reveal social organization in the sleepy lizard. Animal Behaviour 79:217–225.

Leu, S. T., Kappeler, P. M., and C. M. Bull. 2011. Pair-living in the absence of

obligate biparental care in a lizard: trading-off sex and food? Ethology 117:758–

768.

McAlpin, S. 2001. A recovery plan for the Great Desert Skink (Egernia kintorei). Arid

Lands Environment Centre. Alice Springs, Australia.

McAlpin, S. 2011. Social structure and mating system of the Great Desert Skink

Liopholis kintorei. Biological Sciences. Macquarie University, North Ryde, NSW,

Australia.

McAlpin, S., Duckett, P., and A. Stow. 2011. Lizards cooperatively tunnel to construct

a long-term home for family members. PLOS ONE 6:e19041.

Michael, D. R., Cunningham, R. B., and D. B. Lindenmayer. 2010. The social elite:

Habitat heterogeneity, complexity and quality in granite inselbergs influence

! 122! ! ! ! ! ! ! patterns of aggregation in Egernia striolata (Lygosominae: Scincidae). Austral

Ecology 35:862–870.

Morton, S. R., Stafford Smith, D. M., Dickman, C. R., Dunkerley, D. L., Friedel, M. H.,

McAllister, R. R. J., Reid, J.R.W., Roshier, D.A., Smith, M.A., Walsh, F.J.,

Wardle, G.M., Watson, I.W., and M. Westoby. 2011. A fresh framework for the

ecology of arid Australia. Journal of Arid Environments 75:313–329.

Morton, S.R., James, C.D., 1988. The diversity and abundance of lizards in arid

Australia: a new hypothesis. American Naturalist 132:237–256.

Mouton, P. L. F. 2011. Aggregation behaviour of lizards in the arid western regions of

South Africa. African Journal of Herpetology 60:155–170.

O’Connor, D., and R. Shine. 2003. Lizards in “nuclear families”: a novel reptilian

social system in Egernia saxatilis (Scincidae). Molecular Ecology 12:743–752.

O’Connor, D., and R. Shine. 2004. Parental care protects against infanticide in the

lizard Egernia saxatilis (Scincidae). Animal Behaviour 68:1361–1369.

Peakall, R., and P. E. Smouse. 2012. GenALEx 6.5: Genetic analysis in Excel.

Population genetic software for teaching and research-an update. Bioinformatics

28:2537–2539.

Pearson, D., Davies, P., Carnegie, N., and J. Ward. 2001. The great desert skink

(Egernia kintorei) in Western Australia: distribution, reproduction and ethno-

zoological observations. Herpetofauna 31:64–68.

Perrin, N., and V. Mazalov. 2000. Local competition, inbreeding, and the evolution of

sex-biased dispersal. The American Naturalist 155:116–127.

Queller, D. C., and K. F. Goodnight. 1989. Estimating relatedness using genetic

markers. Evolution 43:258–275.

Raymond, M., and F. Rousset. 1995. GENEPOP (version 1.2): population genetics

software for exact tests and ecumenicism. Journal of Heredity 86:248–249.

! 123! ! ! ! ! ! ! Shuttleworth, C., Mouton, P. F. N., and J. H. Van Wyk. 2008. Group size and termite

consumption in the armadillo lizard, Cordylus cataphractus. Amphibia-Reptilia

29:171–176.

Stacey, P. 1986. Group size and foraging efficiency in yellow baboons. Behavioural

Ecology and Sociobiology 18:175–187.

Storr, G. 1968. Revision of the Egernia whiteii species-group (Lacertilia, Scincidae).

Journal of the Royal Society of Western Australia 51:51–62.

Storz, J. 1999. Genetic consequences of mammalian social structure. Journal of

Mammalogy 80:553–569.

Stow, A. J. 2002. Microsatellite loci from the Cunningham’s Skink (Egernia

cunninghami). Molecular Ecology Notes 2:256–257.

Stow, A. J., and P. Sunnucks. 2004. High mate and site fidelity in Cunningham’s

skinks (Egernia cunninghami) in natural and fragmented habitat. Molecular

Ecology 13:419–430.

Stow, A. J., Sunnucks, P., Briscoe, D. A., and M. G. Gardner. 2001. The impact of

habitat fragmentation on dispersal of Cunningham’s skink (Egernia

cunninghami): evidence from allelic and genotypic analyses of microsatellites.

Molecular Ecology 10:867–878.

Sunnucks, P., and D. F. Hales. 1996. Numerous transposed sequences of

mitochondrial cytochrome oxidase I-II in aphids of the genus Sitobion

(Hemiptera: Aphididae). Molecular Biology and Evolution 13:510–524.

Wang, J. 2011. COANCESTRY: a program for simulating, estimating and analysing

relatedness and inbreeding coefficients. Molecular Ecology Resources 11:141–

145.

Watson, J. A., Lendon, C., and B. Low. 1973. Termites in mulga lands. Tropical

Grasslands 7:121–126.

! 124! ! ! ! ! ! ! While, G. M., Uller, T., and E. Wapstra. 2009a. Family conflict and the evolution of

sociality in reptiles. Behavioral Ecology 20:245–250.

While, G. M., Uller, T., and E. Wapstra. 2009b. Within-population variation in social

strategies characterize the social and mating system of an Australian lizard,

Egernia whitii. Austral Ecology 34:938–949.

Wilson, S., and G. Swan. 2010. A Complete Guide to Reptiles of Australia (3rd

Edition). New Holland Publishers.

! ! !

! 125! ! ! ! ! ! ! Supplementary material 4.1

A growth curve for the great desert skink, Liopholis kintorei

Methods

Growth rates were estimated by comparing the SVL of Liopholis kintorei individuals captured one year apart (N = 40). Logistic, Gompertz, and von Bertalanffy growth models were fit to the data using linear regression of the differential form of each equation, as in Kaufmann (1981). The model that yielded the highest R2 value from the linear regression was considered the model that best fitted the data (Kaufmann

1981). The slope of this regression describes the characteristic growth rate (a), or the rate at which the curve approaches the asymptote (Kaufmann, 1981;Johnston,

2011). We found the logistic model to best fit the data:

!!" ! = !!/(1 + !! ) where Sa is asymptotic size, T is time, and L is length (SVL) and B is a parameter calculated from:

! = ! !!/(!! − 1) where L0 is the size at birth (73mm; Pearson et al 2001; equation from Dunham,

1978).

Results

Linear estimates of the differential forms of all growth models tested were significant, however the logistic growth equation provided the best fit to the data (R2 = 0.704, P <

0.0001). Based on the model, adult size (160 mm SVL; Storr 1968) is attained in the third year of growth (Figure S4.1.1), and this was supported by the recapture of two juveniles sampled in the first field season. Separate curves were not calculated for

! 126! ! ! ! ! ! ! males and females due to small sample sizes, and that the sex of immature individuals could not be determined. The estimated model parameters are given in

Table S4.1.1, as well as predicted SVL for a given age.

a' Specific'growth'rate'

Average'SVL'(mm)'

b SVL (mm) SVL

Size at birth (73mm; Pearson et al., 2001)

Age (y)

Fig. S4.1.1. a) Specific growth rates for a given SVL (average calculated from SVL of each individual at T = 0 and after 1 year), and b) logistic growth curve for snout-vent length in Liopholis kintorei, including the literature-derived size at birth (Pearson et al., 2001)

! 127! ! ! ! ! ! ! Table S4.1.1. Growth parameters and predicted SVL (mm) at a given age for Liopholis kintorei.

Asymptotic size (SA), characteristic growth rate (a), and the constant (b) related to size at birth

Age Growth parameters (N = 41) SVL (mm) (years)

SA (mm) 203 1 91 a ± S.E 0.002 ± 0.00015 2 131 b 2.83 3 164 4 184 5 194 6 199

Literature cited

Dunham, A. E. 1978. Food availability as a proximate factor influencing individual

growth rates in the iguanid lizard Sceloporus merriami. Ecology 59:770–778.

Johnston, G. 2011. Growth and survivorship as proximate causes of sexual size

dimorphism in peninsula dragon lizards Ctenophorus fionni. Austral Ecology

36:117–125.

Kaufmann, K. W. 1981. Fitting and using growth curves. Oecologia 49:293–299.

Pearson, D., Davies, P., Carnegie, N., and J. Ward. 2001. The great desert skink

(Egernia kintorei) in Western Australia: distribution, reproduction and ethno-

zoological observations. Herpetofauna 31:64–68.

! ! !

! 128! ! ! ! ! ! ! Supplementary material 4.2

Genetic analyses of within- and among-group relatedness including all animals captured at the main site

Given the low observation rate of L. kintorei, analyses of within-group relatedness and composition were repeated using a less restrictive dataset that included all individuals observed as group members of the burrow system at which they were observed (and not just those captured ≥ 3 times).

Results

Group composition and stability

Using the complete dataset, we identified a total of 25 burrows containing more than one individual in any one season. These ‘groups’ ranged in size from 2 to 9 individuals (mean ± SE = 3.2 ± 0.23), and some individuals were assigned to > 1 group. We assigned 36 individuals (92%; 6 females, 7 males, 23 immatures) to groups in 2011, 43 (90%; 22 females, 13 males, 8 immatures) in 2012, and 28 (90%;

12 females, 7 males, 9 immatures) in 2013.

Group composition varied and membership was dynamic, although some groups had core individuals that remained in the same group across consecutive years. Some individuals were observed as solitary (3 in 2011, 5 in 2012, and 3 in

2013). Given the low observation rate of this species and variability in sampling success, it is possible that individuals categorised as solitary may have been cohabiting with other individuals who were not captured. Similarly, in this less conservative set of analyses (including all animals captured), individuals may have

! 129! ! ! ! ! ! ! been assigned to groups when in fact they were transient visitors to a burrow at which they were captured.

Within seasons, groups were variable, with some individuals being assigned to more than one group within a year (2, 17 and 12 individuals were assigned to > 1 group in 2011, 2012 and 2013, respectively).

Relatedness within and among groups

Average pairwise R between group members (within-groups) was significantly higher than that between non-group members (among-groups) in all three years tested (Table S4.2.1; average within-group relatedness across all years ± SE = 0.227

± 0.02), and this was also the case when only adults were considered. There was insufficient data to test within-group relatedness for females in 2011, however within- group relatedness was higher than among-group relatedness for adult females in both 2012 and 2013, although in 2013 this was not significant (Table S4.2.1).

Although there was insufficient data in 2011 to test males, there was no significant difference in within- and among-group relatedness for adult males in 2012 and 2013.

Immatures from the same group were significantly more highly related than they were to immatures from other groups in all years of sampling.

When relatedness was compared between all group members pooled across years in spatial autocorrelation analyses, 4 out of the 25 groups had significantly higher R than that expected by chance (Figure S4.2.1), however this low number may be due to reduced statistical power because of the low number of pairwise comparisons within groups; within-group relatedness was still relatively high in several groups despite being non-significant (Figure S4.2.1).

! 130! ! ! ! ! ! ! Table S4.2.1. Composition of Liopholis kintorei groups at Newhaven in 2011–2013. The number of individuals belonging to each sex and age class are given for each year of study, as well as the total number of animals captured and average pairwise relatedness for each group (S = solitary)

Group Year Adult females Adult males Immatures Total Within-group R (SE)

B1 2011 - - 2 2 0.7765 (0.00)

2012 3 1 - 4 0.0377 (0.17)

2013 - - - 0 -

B2 2011 - - 3 3 0.1151 (0.15)

2012 - - - 0 -

2013 1 1 - 2 -0.2304 (0.00)

B5 2011 - - 6 6 0.5306 (0.03)

2012 5 1 3 9 0.5411 (0.03)

2013 2 - 3 5 0.5971 (0.03)

B6 2011 1 1 - 2 -0.3007 (0.00)

2012 - - - 0 -

2013 - - - 0 -

B7 2011 - 1 1 2 -0.0785 (0.00)

2012 - - - 0 -

2013 - - - 0 -

B8 2011 1 - 1 2 0.0737 (0.00)

2012 3 1 1 5 0.1845 (0.00)

2013 - - - 0 -

B9 2011 1 1 - 2 -0.1286 (0.00)

2012 - - - 0 -

2013 - - - 0 -

! 131! ! ! ! ! ! ! Group Year Adult females Adult males Immatures Total Within-group R (SE)

B13 2011 1 1 3 5 0.2515 (0.09)

2012 2 2 1 5 -0.0738 (0.08)

2013 1 2 1 4 -0.0569 (0.06)

B14 2011 - 2 2 4 0.2333 (0.10)

2012 - 2 - 2 -0.0408 (0.00)

2013 - - - 0 -

B15 2011 - - - 0 -

2012 1 1 - 2 0.3159 (0.00)

2013 1 1 - 2 0.3159 (0.00)

B16 2011 - 2 - 2 0.5342 (0.00)

2012 1 2 - 3 -0.0519 (0.04)

2013 - - 2 2 0.8989 (0.00)

B17 2011 1 1 - 2 0.6609 (0.00)

2012 2 - - 2 0.5488 (0.00)

2013 1 - 1 2 0.5409 (0.00)

B18 2011 - - - 0 -

2012 1 2 1 4 -0.0773 (0.07)

2013 - - - 0 -

B19 2011 - - - 0 -

2012 1 2 - 3 -0.1151 (0.01)

2013 1 1 - 2 0.0694 (0.00)

B20 2011 - - - 0 -

2012 2 1 1 4 0.4476 (0.07)

2013 2 - 1 3 0.6529 (0.05)

! 132! ! ! ! ! ! ! Group Year Adult females Adult males Immatures Total Within-group R (SE)

B23 2011 - - - 0 -

2012 2 2 - 4 -0.0806 (0.09)

2013 - - - 0 -

B24 2011 1 1 4 6 0.0435 (0.05)

2012 3 1 1 5 0.0151 (0.05)

2013 3 1 1 5 -0.0463 (0.05)

B28 2011 - - - 0 -

2012 - - - 0 -

2013 2 - 1 3 0.0386 (0.03)

B32 2011 - - - 0 -

2012 2 - - 2 0.3038 (0.00)

2013 - - - 0 -

B36 2011 - - - 0 -

2012 2 2 - 4 -0.0652 (0.04)

2013 1 2 - 3 0.1127 (0.08)

B43 2011 - - - 0 -

2012 1 1 - 2 -0.1839 (0.00)

2013 1 1 - 2 0.5235 (0.00)

B44 2011 - - - 0 -

2012 - - - 0 -

2013 1 1 - 2 0.0601 (0.00)

B45 2011 - - - 0 -

2012 1 - 2 3 0.4233 (0.02)

2013 1 - 2 3 0.4233 (0.02)

! 133! ! ! ! ! ! ! Group Year Adult females Adult males Immatures Total Within-group R (SE)

B48 2011 - - - 0 -

2012 1 1 - 2 -0.0993 (0.00)

2013 - - - 0 -

B52 2011 - - - 0 -

2012 - - - 0 -

2013 1 1 - 2 -0.0377 (0.00)

S 2011 - 1 2 3 -

2012 3 2 - 5 -

2013 2 1 - 3 -

! 134! ! ! ! ! ! ! Table S4.2.2. Average relatedness (R) between Liopholis kintorei within and among social groups at

Newhaven each year. Standard errors (SE) and P-values were calculated via bootstrapping. In all statistically significant cases, within-group relatedness was higher than among-group relatedness. “na”! indicates insufficient data for comparisons

! Year! R! SE! ! R! SE! P*value!

! " " " " " " " All!adults!&!immatures! 2011! 0.2573! 0.0414! ! .0.0105! 0.0073!

! 2012! 0.2172! 0.0348! ! .0.0297! 0.0064!

! 2013! 0.2077! 0.0517! ! .0.0541! 0.0096!

! ! ! ! ! ! ! ! Adults!only! 2011! 0.0847! 0.1520! ! 0.0279! 0.0200! NS!

! 2012! 0.1265! 0.0414! ! .0.0257! 0.0075!

! 2013! 0.0285! 0.0583! ! .0.0459! 0.0144! NS!

! ! ! ! ! ! ! ! Adult!females! 2011! na! na! ! na! na! na!!

! 2012! 0.3020! 0.0690! ! .0.0341! 0.0125!

! 2013! 0.0458! 0.1660! ! .0.0779! 0.0212! NS!

! ! ! ! ! ! ! ! Adult!males! 2011! na! na! ! na! na! na!!

! 2012! .0.0678! 0.0532! ! 0.0062! 0.0220! NS!

! 2013! .0.0141! 0.0185! ! 0.0473! 0.0426! NS!

! ! ! ! ! ! ! ! Immatures! 2011! 0.3623! 0.0502! ! .0.0459! 0.0117!

! 2012! 0.6574! 0.0517! ! .0.0411! 0.0375!

! 2013! 0.5976! 0.0714! ! .0.0387! 0.0239!

! 135! !

0.800%

0.600%

0.400%

0.200%

0.000% relatedness)

!0.200%

!0.400%

!0.600%

136 B1% B2% B5% B6% B7% B8% B9% B13% B14% B15% B16% B17% B18% B19% B20% B23% B24% B28% B32% B36% B43% B44% B45% B48% B52% Group) !

Fig. S4.2.1. Average pairwise relatedness between group members pooled across years. Red marks represent the 95% confidence interval around which

relatedness is effectively zero

! CHAPTER 5

Do bigger heads make burly males? Sexual selection and the

evolution of head size dimorphism in lizards

Photo by Siobhan Dennison

The following manuscript is currently in preparation for submission to the Journal of Zoology. It is presented in the format required for this journal with the exception of table and figure numbers. These have been altered to read sequentially throughout this thesis.

! 137! Abstract

Sexual size dimorphism (SSD) is a widespread phenomenon in which differential selection pressures result in differences in morphometric traits for each sex. In lizards, sexual selection can favour larger size in males because of the advantages conferred during male–male competition for territory or mates. Comparative studies to date have assessed sexual dimorphism in body size in this context. However, simultaneous selection for increased fecundity in larger females may obscure the intensity of selection acting upon males. I therefore set out to investigate sexual selection for sexual dimorphism in head size (SDHS), with the prediction that SDHS will be associated with the degree of polygyny. Initially I tested this prediction in the

Egernia group of scincid lizards, which has species exhibiting a range of mating systems, from monogamy through to high levels of polygyny. I then extended this investigation across lizard families (60 species, 39 genera and 15 families), using territoriality as a proxy for polygyny. Here I reviewed the available literature on sex- specific morphometrics, to test whether more territorial species exhibited more extreme SDHS. I found no significant difference in sexual head size dimorphism between different categories of territorial behaviour, and discuss possible explanations for this result.

Introduction

Sexual size dimorphism (SSD), where one sex is larger than the other in one or more morphological traits, is present in many animal species, however the direction and degree to which it occurs varies widely. In many vertebrate taxa including birds, squamate reptiles and mammals, males are often the larger sex (e.g. Fairbairn and

Shine, 1993; Weckerly, 1998; Cox et al., 2003), while females are larger in many fish,

! 138! birds of prey and amphibians (e.g. Andersson and Norberg, 1981; Bisazza, 1993;

Shine, 1979).

Numerous mechanisms have been proposed to explain the evolution of SSD.

These patterns may be influenced by natural selection, for example, larger-bodied females may have an advantage because of enhanced fecundity, because a larger size allows them to produce larger clutches or larger-sized offspring (Madsen, 1983;

Shine, 1994). Intersexual niche separation may also drive SSD because it can reduce competition for resources, particularly food, between the sexes (Madsen,

1983; Shine, 1989). On the other hand, sexual selection can play an important role.

Sexual selection is the result of variance in mating success, and from a male perspective, selection will favour those traits that maximise the number of matings he can attain (Uller and Olsson, 2008). For example, larger body size may increase a male’s reproductive success if it increases his ability to win contests for access to females or territories (Cox et al., 2007; Shine, 1978).

Consequently, if sexual selection acts upon SSD, one may therefore expect that SSD is more pronounced in polygynous than in monogamous species (Stamps, 1983, but see Bull and Pamula, 1996).

Squamate reptiles (lizards and snakes) exhibit a wide range of sexual dimorphism, which reflects the diversity of life history strategies, and mating and social systems, within the group. In lizards, larger males are more likely to own territories, with territory owners having greater access to females (Stamps, 1983).

Dominant males that can hold large or high quality territories encompassing the home ranges of multiple females have higher mating success than subordinate conspecifics (Stamps, 1983; Hews, 1993).

The majority of studies investigating sexual size dimorphism in reptiles have focused on differences in body size (e.g. Shine 1978, 1994; Cox et al., 2003),

! 139! however, in numerous species sexes do not differ in overall body size, but males have proportionally larger heads (e.g. Bull and Pamula, 1996; Dubey et al., 2011).

Male head size is thought to be under selection because males with larger heads

(and correspondingly stronger jaws; Carothers, 1984; Lappin et al., 2006) are more successful in contests.

Given that males often have larger heads than females, and that selection may be simultaneously favouring larger body size in females, standard measurements of SVL may not reflect any net differences between the sexes. We may therefore expect to see a stronger relationship between estimators of intrasexual selection intensity and SSD in head size (SDHS) compared with SSD in body size.

To test the hypothesis that male-biased sexual head size dimorphism is associated with the degree of polygyny in lizards, I measured the association between mating system and relative head size data in the Egernia group. The

Egernia group was selected because social organisation ranges from solitary individuals to stable social group living, and mating systems are characterised by long-term monogamy through various degrees of polygyny (Bull, 2000; Chapple,

2003; Stow and Sunnucks, 2004; Chapple and Keogh, 2005; While et al., 2009b;

McAlpin et al., 2011). Several species live in groups of highly-related individuals, often including an adult pair and offspring from multiple age cohorts (Gardner et al.,

2001; O’Connor and Shine, 2003; Stow and Sunnucks, 2004; Chapple and Keogh,

2005), and individuals have been known to defend core territories from unrelated conspecifics (Sinn et al., 2008; While et al., 2009a). Given an association between polygyny and SSD (Stamps, 1983), I expected to see corresponding variation in the degree of SSD within the Egernia group.

Although not well documented within Egernia, SSD appears in a number of species within the group. In Egernia major, males and females do not differ in SVL,

! 140! however males have larger heads than females of the same SVL (Osterwalder et al.,

2004). Similarly, Clemann et al., (2004) found that male E. coventryi have larger heads relative to SVL than females, as do male Liopholis whitii (Chapple, 2005).

Female L. whitii have also been found to have larger body size than males, and females of both L. whitii and E. stokesii with larger SVL also had higher fecundity

(Chapple, 2005), indicating a selective advantage for increased female body size in these species. Although no studies in Egernia have explicitly tested whether variance in size is related to variance in male reproductive success (which provides opportunity for sexual selection), this has been shown previously in other lizard species (e.g. Censky 1995; Marco and Perez-Mellado, 1999; Lebas 2001). Further, the ability to acquire multiple social partners and avoid cuckolding has been shown to increase variation in male reproductive success in Liopholis whitii (While et al., 2011).

Acquisition of multiple females is likely to be driven by territorial behaviours, which implies strong selection on traits that increase the quality and size of a male’s territory, such as aggressive behaviour and body or head size (Olsson and Madsen

1998; While et al., 2011), however this has not yet been documented in Egernia, and further work is required to paint a more complete picture of the influence that sexual selection has on variation in sexual traits within the lineage.

One species within the Egernia group, Liopholis kintorei, excavates extensive tunnel systems in which close kin live. The tunnels offer protection from aerial predators and thermal extremes in the desert, as well as a shelter site in which the young mature (McAlpin et al., 2011; Fenner et al., 2012). Their construction also represents a significant investment of time and energy, and given the long-term occupancy of these burrows it is likely that, like other members of the Egernia group

(e.g. E. saxatilis – O’Connor and Shine, 2004; L. whitii – While et al., 2009a), they defend their shelter sites from conspecifics. Of the group-living Egernia species, L.

! 141! kintorei has the highest level of genetic polygyny recorded in Egernia (40% of males mated with multiple females at Uluru; McAlpin et al., 2011), and males may sire young within several burrow complexes in close proximity (McAlpin et al., 2011).

However, in another study of a geographically and genetically differentiated population at Newhaven, NT (S Dennison, unpubl. data; Chapter 4 this thesis), no male L. kintorei were found to have multiply mated. Within-group relatedness was significantly higher than that among groups (as was the case at Uluru, McAlpin et al.,

2011), and the presence of full-sib adult daughters in burrows indicates some monogamy, however only 7 mated pairs were identified in that study, which may not be representative of the population. Males at Newhaven were also found to move among burrows more than females (Chapter 3 this thesis), presumably looking for extra-pair mating opportunities, and were often found in burrows with multiple females. Given the high intraspecific variation in mating systems within Egernia species (e.g. Chapple and Keogh, 2005; While et al., 2009b) the presence of genetically polygynous males cannot be ruled out. It is likely that in L. kintorei, their home-sites and the females within them constitute the basis for resource-defense polygyny, providing a selective pressure for SSD in this species.

Here I examined the hypothesis that sexual selection acts upon SSD in head size in Egernia group species, by comparing the extent of SDHS and degree of genetic polygyny. Given that McAlpin et al., (2011) found the highest degree of genetic polygyny known in the group-living Egernia species in L. kintorei, I included data from both Uluru (McAlpin 2011), as well as Newhaven (where no polygyny was detected; Chapter 4 this thesis) in order to compare it to its congeners. I then extended this to test the sexual selection hypothesis for SDHS across a taxonomically diverse range of lizards by reviewing available literature on sex- specific morphometrics across lizard families. There were not sufficient data in the

! 142! literature that directly quantified polygyny to enable its use as a measure of intrasexual selection. The vast majority of genetic studies I found assessed multiple mating from the female perspective (i.e. multiple paternity in clutches; Uller and

Olsson, 2008, and references therein), but not that of males. Levels of multiple mating by males are rarely directly quantified, but are often estimated using other variables such as the ratio of male/female territory size (Stamps, 1983; Cox et al.,

2003). I used territoriality as a measure of the intensity of intrasexual selection because it increases the variance in male mating success through increased access to females (reviewed in Stamps, 1983; Anderson and Vitt, 1990; Hews, 1990; Cox et al., 2003), and has been shown to be associated with polygyny and SSD in lizards

(reviewed in Stamps, 1983; Cox et al., 2003). I predicted that in more territorial species, the extent of SSD would be greater than in non-territorial species.

Methods

Sexual dimorphism and polygyny in Egernia

I exhaustively searched the literature on Egernia group species for data on sex- specific morphometrics and polygyny. Because of the paucity of data on male multiple mating, I also used within-group relatedness as a proxy for the degree of polygyny, because this should increase as the level of multiple mating decreases

(Stow and Sunnucks, 2004).

To test for SDHS in Liopholis kintorei, individuals were sampled from the

Australian Wildlife Conservancy’s Newhaven Sanctuary (22º 48’ S, 131º 15’ E), approximately 360 km northwest of Alice Springs, Northern Territory, Australia.

Lizards were captured in Elliot traps at a 12 ha study site which was intensively sampled during the breeding season across three consecutive years (2011–2013).

Full capture methods are outlined elsewhere (Chapters 3 and 4 this thesis). In

! 143! addition to the main study site, I also trapped lizards at a number of additional locations throughout the Newhaven property in 2012. For adult lizards captured in consecutive years, I used the measurements taken in 2012 because sampling success was highest in this year. Linear measurements of head dimensions were taken to the nearest 0.1 mm with digital calipers. I measured head length (HL) from the anterior edge of the right auricular opening to the tip of the snout, head width

(HW) between the auricular openings, and head height (HH) at the highest point of the head, just posterior to the orbita. SVL was measured from the tip of the snout to the posterior edge of the cloacal aperture with a ruler to the nearest millimetre. Sex was determined by hemipene eversion, and recorded for all mature adults (>160 mm;

Storr 1968).

I first tested for significant differences in SVL between males and females using ANOVA. Differences in head size between the sexes were tested using

ANCOVA with HW, HL and HH as dependent variables and SVL as the covariate. In addition to this, because HW, HL and HH are highly correlated (see Table S5.1,

Supplementary material 5.1), I also calculated a measure of relative head size for every individual (RHS; see below for calculation), and compared this between males and females using ANOVA.

No comparative statistical analysis could be carried out across the Egernia group species due to the small data set, and so I simply report trends in the Results.

Sexual dimorphism and territoriality in lizards

Sex-specific data on head size relative to SVL were collected for as many lizard species as possible using literature searches on ISI Web of Science (search terms: sexual dimorphism, head size, reptile, lizard, morphometric), and by screening the

! 144! literature cited in these papers. To calculate the extent of SDHS, I first calculated an index of relative head-size (RHS) for each sex:

(HW × HL) RHS = SVL

I then used an equation from Smith (1999), modified from Lovich and Gibbons (1992) to determine the extent of SSD for each species:

size of males SSD!=! size of females

This measure of SSD is considered to have the advantage over a simple ratio in that it is symmetrical around the value of zero (Lovich and Gibbons, 1992), and is considered appropriate when species in which the males are larger dominate the dataset, as was the case in this study. Territoriality was scored for each species according to the three categories defined by Stamps (1977): (1) non-territorial – the species does not defend any part of its home range; (2) specific site defense – defense of a small site within the home range (e.g. a perch, shelter or basking rock); and (3) home range defense – active defense of most or all of the home range.

Observations of space use and territoriality in lizards can range from quantitative measures (e.g. home range size or overlap) to short descriptive statements, which can often be rather subjective. If a direct statement of territoriality status could not be found in the literature, an estimate was obtained based on one of several additional criteria. If a species was described simply as ‘territorial’, or male– male aggression observed with no elaboration, the species was scored according to the most common categorisation for its family (either home-range or specific-site

! 145! defense). According to Stamps (1977), widely-foraging species with extensive home

ranges are usually considered to be non-territorial. For the few species described in

this way, I assigned them as non-territorial according to this criterion. Any species for

which I had morphometric data that could not be assigned to one of the three

categories of territoriality were excluded from further analysis, and any species for

which phylogenetic data were not available (sensu Pyron et al., 2013) were also

excluded from the analysis.

I first tested for phylogenetic signal in the data using Moran’s I autocorrelation

index (Gittleman and Kot, 1990). I then used a phylogenetic generalised least

squares (PGLS) model to assess the relationship between the extent of SDHS

(dependent variable) and the level of territoriality (predictor), using the package lme4

(Bates, 2005), implemented in R 3.1 (R development core team 2013; www.r-

project.org). To account for taxonomic bias in both SSD and territoriality,

phylogenetic correlation was incorporated into the model as a covariate using

phylogenetic distances calculated by Pyron et al., (2013). Because data from more

than one species were sometimes obtained from the same study, I also included

‘study’ as a random effect to account for any associated sampling bias.

Results

Male-biased sexual head size dimorphism in L. kintorei

I captured 51 adult female and 29 adult male L. kintorei. I found no significant

difference in SVL between the sexes (F1,79 = 3.33, P = 0.072), however males had

larger head dimensions relative to body size than females (HW – F1,79 = 4.84, P =

0.031; HL – F1,79 = 27.71, P < 0.001; HH – F1,79 = 8.87, P = 0.004; RHS – F1,79 =

14.39, P < 0.001; Table 5.1).

! 146! Table 5.1. Summary of body and head measurements (mm) for male and female L. kintorei

Adult male (n=29) Adult female (n=51)

Trait Mean ± SE Range Mean ± SE Range

Head width 27.87 ± 0.44 23.44–30.83 26.54 ± 0.30 21.58–29.85

Head length 34.94 ± 0.46 30.69–40.07 32.92 ± 0.28 28.61–37.26

Head height 25.71 ± 0.55 20.13–31.17 23.85 ± 0.29 19.40–28.45

Snout-vent length 188.59 ± 2.37 160–211 183.29 ± 1.72 160–204

Relative head size 5.15 ± 0.10 4.10–5.96 4.77 ± 0.05 3.92–5.60

Polygyny as a predictor of SDHS in Egernia

In addition to L. kintorei at Newhaven (NH), data were available to calculate SDHS

for another population of this species at Uluru (UL), and another three species from

the Egernia group: L. whitii, E. striolata and E. cunninghami (Table 5.2). Data on

within-group R were available from all of these species, and levels of polygyny were

available for all except E. striolata. Overall, the paucity of data precluded any

statistical analysis, however there was a slight trend for the extent of SDHS to

increase as within-group R increased (Figure 5.1), and there was no visible

relationship between SDHS and the degree of polygyny (Figure 5.2). More data

would be required to investigate these relationships further.

Table 5.2. Data collected for Egernia group species: the extent of SDHS, within-group relatedness,

degree of polygyny (% males that multiply mated), and references are given for each species. “na”

indicates that data were not available for a given variable

Within- % polygynous Species SDHS References group R males Egernia cunninghami 0.069 0.264 13 Stow et al., 2001 Egernia striolata 0.097 0.208 na Duckett et al., 2012 Liopholis whitii 0.099 0.221 33 Chapple and Keogh, 2005 Liopholis kintorei (NH) 0.105 0.287 0 S. Dennison (Chapter 4) Liopholis kintorei (UL) 0.236 0.371 40 McAlpin et al., 2011

! 147! 0.250$

L.#kintorei#(UL)# 0.200$

0.150$ Extent&of&SDHS& 0.100$ L.#whi3i# L.#kintorei#(NH)# E.#striolata# E.#cunninghami# 0.050$ 0$ 0.05$ 0.1$ 0.15$ 0.2$ 0.25$ 0.3$ 0.35$ 0.4$ Within/group&relatedness&(R)&

Fig. 5.1. Slight positive trend for the extent of SDHS to increase with within-group R in species of the

Egernia group

0.250$ L.#kintorei#(UL)#

0.200$

0.150$ Extent&of&SDHS& 0.100$ L.#kintorei#(NH)# L.#whi.i#

E.#cunninghami# 0.050$ 0$ 10$ 20$ 30$ 40$ 50$ %&polygynous&males&

Fig. 5.2. No relationship was evident between the extent of SDHS and levels of polygyny in species of the Egernia group

! 148! Territoriality as a predictor of SDHS in lizards

My data set included 60 species from 39 genera and 15 families for which SDHS could be calculated and an estimate of territoriality obtained (Table 5.3). There was no significant difference in SDHS among the three territoriality categories (ANOVA –

F2,57 = 0.012, P = 0.988; Figure 5.3).

There was a significant effect of phylogeny on both SDHS (Moran’s I = -0.036,

P < 0.001) and territoriality (Moran’s I = -0.093, P < 0.001). Overall in the PGLS model, after accounting for phylogenetic effects, the level of SDHS could not be predicted by the level of territoriality (Figure 5.3; Table 5.4). The random effect

(‘study’) had a no significant effect on the model (P = 0.868; Table 5.4).

! 149! ' ! ! ! ! ! ! ! ! ! ; see ! ! ! ! Harlow!and!Taylor,! 2000 Ji!et!al.,!2002 Doughty!et!al.,!2007 Shine!et!al.,!1998 Du!et!al.,!2011 Qi!et!al.,!2012 Cordes!et!al.,!1995 Cordes!et!al.,!1995 Mouton!et!al.,!1999 Mouton!and!van!Wyk,! 1993 Herrel!et!al.,!2007 References'(Morph.) Reaney!and!Whiting,! 2002 Znari!and!El!Mouden,! 1997 ! ! ! ! ! ' ;! ! ! ! El!Mouden!et! ! ! ! ' ;! rlow!and!Taylor,! al.,!1999 Ha 2000 Radder!et!al.,!2001 Healey!et!al.,!2007;! Olsson!et!al.,!2007 McGuire!and!Dudley,! 2011 Du!et!al.,!2011 Qi!et!al.,!2012 Cordes!et!al.,!1995 Cordes!et!al.,!1995 Mouton!et!al.,!1999 Mouton!and!van!Wyk,! 1993 Schoener!and! Schoener,!1982 Edwards!and!Lailvaux,! References'(T) Reaney!and!Whiting,! 2003 Znari!and!El!Mouden,! 1997 vent length (SVL) are given in mm in given are (SVL) length vent icate species in which females are - ' ! ! ! ! ! ! ! ! ! ! ! ! ! 1.15 1.11 1.18 0.91 1.17 1.04 1.15 1.20 1.16 1.49 1.31 1.26 1.16 SDHS e defense, 3 = home range defense/highly defense/highly homerange = 3 defense, e ! ! ! ! ! ! ! ! ! ! ! ! ! ' sit - RHS 6.07 3.03 3.06 1.75 2.85 4.07 4.52 4.53 7.96 5.25 2.26 6.85 5.58 ! ! ! ! ! ! ! ! ! ! ! ! ' ! ' 95 SVL 90.5 89.8 60.3 77.9 55.4 72.4 74.4 101.4 105.6 55.36 117.02 105.77 ! ! ! ! ! ! ! ! ! ! ! ! ! ' FEMALES HL 20.3 28.9 26.1 29.2 19.6 15.9 14.5 19.8 19.5 15.27 27.04 29.68 15.85 territorial, 2 = specific - ! ! ! ! ! ! ! ! ! ! ! ' ! ! 8.2 9.4 HW 16.6 29.1 19.1 18.8 13.9 11.6 14.6 16.8 29.63 19.88 14.24 ! ! ! ! ! ! ! ! ! ! ! ! ! ' RHS 5.44 9.27 7.83 2.96 8.60 6.49 7.00 3.37 3.60 1.60 3.35 4.25 5.19 ! ! ! ! ! ! ! ! ! ! ! ! ! ' ' SVL 106 78.8 91.7 90.1 59.7 75.7 57.2 72.5 111.2 62.58 111.5 117.22 113.09 ! ! ! ! ! ! ! ! ! ! ! ! MALES ' ! 32 HL 22.8 33.2 31.3 20.5 17.2 13.6 22.5 20.8 18.81 28.71 34.17 16.64 ! ! ! ! ! ! ! ! ! ! ! ' ! ! 25 8.9 HW 18.8 32.2 9.84 20.5 14.8 12.5 16.6 18.1 35.12 21.49 14.62 ! ! ! ! ! ! ! ! ! ! ! ! ! ' * ## ## T 2 2 3 3 3 3 3 3 2 2 2 3 3 , , , ! , , , , , ! , ! , , relativehead size (RHS; see text for calculation). The extent of sexual head size dimorphism isgiven for each species (SDHS '

Draco,melanopogon Leiolepis,reevesii Phrynocephalus,vlangalii Cordylus,cordylus Cordylus,niger Ouroborus,cataphractus Pseudocordylus,melanotus Anolis,carolinensis Species Acanthocercus,atricollis Agama,impalearis Amphibolurus,muricatus Calotes,versicolor Ctenophorus,pictus Species list (by family), data and references used in this study. Mean head width (HW), head length (HL), snout (HL), length head (HW), width head Mean study. this in used references and data family), (by list Species .3. 5 ' ' ' Table Table for each sex, as well as text for calculation), as is the level of territoriality (T, as definedin Stamps 1977; 1 = non the largersex territorial). References are given separately for territoriality (T) and morphometric data (Morph.). Negative SDHS values ind ' Dactyloidae Cordylidae Agamidae Family

150 ! ! ' ! ! ! ! ! ! ! ! ! ! ! onner,!1996 ! ! ! ! ! aña,!1996 Herrel!et!al.,!1999 Braña,!1996 Br Braña,!1996 van!der!Meer!et!al,! 2010 van!der!Meer!et!al,! 2010 Braña,!1996 Braña,!1996 Braña,!1996 Huang,!1998 Huang,!2006 References'(Morph.) Perry,!1996 Vitt,!1986 Saenz!and!C Whiting!et!al.,!2007 Vitt,!1986 Gienger!and!Beck,!2007 Lappin!et!al,!2006 ! ^ ' ! ! ! ;!refs! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Borja,!1985;!Herrel!et! al.,!1999 Galán,!2008 Marco!and!Perez Mellado,!1999 Stamps,!1977 therein M.!Whiting,!pers. comm. M.!Whiting,!pers. comm. Galán,!2000 Carazo!et!al.,!2011;! López!and!Martín,! 2001 Font!et!al.,!2012 Huang,!1998 Kikukawa!and!Hikida,! 2012 References'(T) 2013 Perry,!1996 Vitt,!1986 Paulissen!et!al.,!2013 M.!Whiting,!pers. comm. Vitt,!1986 Beck!and!Jennings,! 2003 Kwiatkowski!and! Sullivan,!2002 ' ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 1.43 1.39 1.41 1.44 1.02 1.01 1.43 1.36 1.40 1.36 1.07 1.13 1.08 1.07 1.01 1.04 1.16 1.57 SDHS ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' RHS 3.35 2.79 5.11 4.88 1.67 1.66 2.35 2.06 2.44 1.84 1.61 2.05 2.80 2.30 5.10 1.34 7.06 3.19 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ! ' 71 SVL 57.1 61.2 56.6 51.5 29.7 93.33 62.73 101.9 89.11 53.51 47.63 67.08 45.16 94.57 244.4 151.3 108.54 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' FEMALES HL 14.1 12.2 25.5 7.38 45.9 26.2 19.6 33.1 15.3 17.69 15.29 13.28 14.74 22.13 32.33 14.21 13.01 17.08 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ! ! ! 8 9.7 6.4 HW 7.87 7.07 6.96 5.39 37.6 18.4 8.94 9.04 7.37 10.77 18.92 14.13 16.12 15.99 10.49 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' RHS 3.42 2.50 1.73 2.31 3.03 2.46 5.14 1.39 8.22 4.99 4.79 3.89 7.23 7.02 1.71 1.67 3.37 2.80 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ! ! ' 60 28 SVL 64.2 59.2 50.9 55.3 57.23 49.51 92.31 249.9 168.9 61.13 80.57 69.08 50.83 109.65 102.31 107.15 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! MALES ' ! 49 HL 15.8 15.6 12.4 7.27 32.8 20.52 14.75 15.47 25.28 28.18 22.74 36.66 37.55 13.68 13.25 20.25 18.11 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ! 9.5 HW 9.54 7.12 7.74 10.1 5.37 41.9 25.7 8.72 9.21 7.85 11.59 18.77 18.62 10.46 20.17 20.03 10.06 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' † † † T 3 1 1 3 3 1 2 3 3 1 3 1 1 1 2 1 1 1 , , , , , , , , , , , , , , , , , , bocagei ' acerta,monticola Nucras,lalandii Nucras,tessellata Podarcis, Podarcis,hispanica Podarcis,muralis Takydromus,hsuehshanensis Takydromus,sauteri Lygodactylus,klugei Heloderma,suspectum Sauromalus,ater Gallotia,galloti Iberol Lacerta,schreiberi Lacerta,viridis Species Anolis,polylepis Hemidactylus,mabouia Hemidactylus,turcicus Homopholis,walbergii ' ' ' (cont.) ' ' ' ' Lacertidae Iguanidae Lacertidae Gekkonidae Helodermatidae Family

151 ' ! ! ! ! ! ! ! ! ! ! ! ubl.!data p ! ! ! Salinas!et! ^ ! ! ! ! 2010 Noble,!un Vitt!and!Cooper,!1985 Vitt!and!Cooper,!1985 Cabrera!et!al.,!2013 Cabrera!et!al.,!2013 Cabrera!et!al.,!2013 Randriamahazo,!2000 Hernández al.,! Hews,!1996 Vitt,!1986 D. S. Dennison!(this!study) Chapple!and!Keogh,! 2005 ! ! References'(Morph.) Luo!et!al.,!2012 Braña,!1996 Braña,!1996 Gifford!and!Powell,! 2007 Cabrera!et!al.,!2013 Cabrera!et!al.,!2013 Cabrera!et!al.,!2013 Cabrera!et!al.,!2013 ! ! ! ! ! ! ! ! ! ! ! ! ' ! ! ! ! ! ! ! tins!et!al.,!2004 tins!et!al.,!2004 Martins!et!al.,!2004;! Robles!and!Halloy,! 2010 Mar Martins!et!al.,!2004 Randriamahazo,!2000 Stamps,!1977 Hews,!1990 Vitt,!1986 Carazo!et!al.,!2014 McAlpin!et!al,!2011 While!et!al.,!2009 Bateson!et!al.,!2011 Cooper!and!Vitt,! 1987;!Griffith,!1991 References'(T) Kikukawa!and!Hikida,! 2012 Castilla!and!Bauwens,! 1990 Stamps!1983 Jenssen!et!al.,!1989 Martins!et!al.,!2004 Martins!et!al.,!2004 Martins!et!al.,!2004 Mar ' ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 1.17 1.23 1.21 1.21 1.17 1.22 1.03 1.20 1.11 1.10 1.30 1.28 1.06 1.39 1.26 1.28 1.15 1.08 1.24 1.09 SDHS ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' RHS 1.99 2.53 2.57 4.14 2.11 2.40 3.68 2.84 4.81 2.83 1.84 1.86 1.24 7.35 1.99 2.29 2.32 1.65 2.29 2.12 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ' SVL 63.5 75.5 86.7 63.3 66.1 48.7 45.9 51.45 58.19 55.03 132.3 49.32 53.64 60.61 51.27 56.35 52.11 105.86 184.75 135.29 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ! FEMALES 13 HL 9.9 20.2 12.7 15.1 11.5 26.6 21.65 33.13 12.91 44.65 13.28 11.54 10.16 12.31 11.65 13.56 12.89 11.16 18.61 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ! ! 9.5 6.1 HW 13.9 9.19 7.07 7.44 10.5 9.47 8.92 20.6 9.33 11.7 26.82 12.16 22.27 10.44 10.32 10.86 10.97 14.93 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' RHS 3.42 5.32 3.11 2.39 2.37 1.32 2.51 2.93 2.67 1.78 2.84 2.30 2.34 3.10 3.12 5.00 2.48 2.92 3.79 10.25 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ! ' 74 SVL 83.6 63.1 71.9 46.3 74.9 47.91 76.54 52.88 42.98 60.25 52.59 54.66 69.13 63.81 153.8 51.99 106.41 190.46 140.78 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! MALES ' ! 10 HL 31.8 23.33 35.47 20.93 14.14 14.98 51.02 15.75 16.73 12.63 10.22 14.16 12.12 12.89 16.05 15.61 12.39 15.38 18.62 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ' ! ! 6.1 7.5 HW 7.64 9.98 9.92 24.2 15.58 28.54 12.44 10.66 11.38 28.27 13.39 11.16 12.08 13.37 12.76 10.39 14.04 15.26 ! ! * * T' 3! 1! 1! 1! 3! 1! 3! 3! 3! 3! 3! 3! 3! 3! 3! 3! 3! 3! 2 2 , , , , , , , , , , , , , , , , , , , , vivipara , ' lepida , Oplurus,cuvieri Sceloporus,grammicus Uta,palmeri Phyllopezus,pollicaris Eulamprus,quoyii Liopholis,kintorei Liopholis,whitii Plestiodon,fasciatus Plestiodon,inexpectatus Liolaemus,abaucan Liolaemus,chacoensis Liolaemus,koslowskyi Liolaemus,laurenti Liolaemus,quilmes Liolaemus,salincola Liolaemus,scapularis Species Takydromus,wolteri Timon Zootoca Leiocephalus,schreibersii ' ' ' ' ' ' ' pluridae Phyllodactylidae Scincidae O Phrynosomatidae Leiocephalidae Liolaemidae Family

15 ' ! ! ! ! ! ! ! ! ! and!Cooper,!1985 tt!and!Zani,!1996 Vitt! Pinto!et!al.,!2005 References'(Morph.) ! Becker!and!Paulissen,! 2012 Vitt,!1983 Vitt,!1991 Vitt!et!al.,!1995 Vitt!and!de!Carvalho,! 1992 Vitt,!1991 Vi ! ! ! ' ! ! ! ! ! 01 Ribeiro!et! ranges ; ! Ribeiro!et!al.,!2009 References'(T) Stamps,!1977;!Vitt! and!Cooper,!1985 Stamps,!1977 Vitt,!1983;! al.,!2011 Vitt!et!al,!1995 Vitt!et!al,!1995 Vitt!et!al,!1995 Kohlsdorf,!20 Van!Sluys!et!al.,!2004 ' ! ! ! ! ! ! ! ! ! 1.37 1.68 1.11 1.14 1.08 1.27 1.15 1.28 1.42 SDHS ! ! ! ! ! ! ! ! ! ' RHS 3.67 2.98 0.83 2.20 3.02 3.53 2.67 5.43 3.39 ! ! ! ! ! ! ! ! ! ' ' SVL 104 94.8 45.5 83.1 87.57 55.47 80.76 114.9 90.11 ! ! ! ! ! ! ! ! ' ! FEMALES 24 HL 7.46 19.5 19.3 25.9 19.6 19.25 15.31 19.73 ! ! ! ! ! ! ! ! ! ' in!this!family HW 5.09 7.96 12.5 15.3 14.69 11.49 24.11 15.57 16.31 ! ! ! ! ! ! ! ! ! ' RHS 5.00 0.92 2.51 3.27 4.50 3.07 6.94 4.82 5.04 territoriality!in!widely!foraging!species!with!extensive!home ; ! ! ! ! ! ! ! ! ' ! territoriality!is!more!common!in!this!family ! ' 89 SVL 106 42.3 79.8 109.4 58.88 139.8 107.27 101.86 ! ! ! ! ! ! ! ! ! MALES defense ' HL 7.56 26.5 21.3 24.3 25.85 16.71 20.06 32.34 24.62 territoriality!is!more!common! ! range! ! ! ! ! ! ! ; ! ! ' ! 18 HW 5.17 8.85 21.16 13.02 12.81 30.01 20.98 21.13 defense ! ! ! ! † † † # T' 2! 2! 1! 3! 3! 1 1 1 3 range! ; r!sites!that!are!regularly!utilised!by!group!members , , , , , , , , , ' territorial!based!on!Stamps’!(1977)!description!of!non ; Kentropyx,pelviceps Kentropyx,striatus Plica,plica Tropidurus,flaviceps Tropidurus,torquatus Species Plestiodon,laticeps Scincella,lateralis Cnemidophorus,ocellifer Kentropyx,calcarata male!combat!recorded!and!home – ' described!only!as!‘territorial’!in!literature!and!home male classified!as!non group!living!species,!but!defend!shelte !! !!!! !!!! # ## † * ' ' Tropiduridae Teiidae Family

15 0.3"

0.2"

0.1" Extent&of&SSD&(ra.o)&

Non-territorial Specific site defense Territorial 0"

Fig. 5.3. Extent of sexual head size dimorphism (± SE) in species categorised as: non-territorial, specific site defense, territorial (home-range defense). See text for calculations

Table 5.4. Phylogenetic generalised least-squares model and parameter estimates for the prediction of the extent of SDHS based on level of territoriality

Source of variation Coefficient (± SE) t-statistic P-value intercept 1.129 ± 0.121 9.353 0.000

Territoriality 0.020 ± 0.022 0.909 0.367

Study (random effect) 0.000 ± 0.001 0.173 0.864

Discussion SDHS and polygyny in Egernia

Male-biased dimorphism in head size in the great desert skink, L. kintorei, from both

Uluru and Newhaven was higher than the other group-living Egernia species in this study. Liopholis kintorei at Uluru had a higher degree of dimorphism than those at

Newhaven, which is consistent with the disparity in levels of genetic polygyny recorded for these locations (40% at Uluru – McAlpin et al., 2011; none identified at

Newhaven – Chapter 4 this thesis), however, no trend was apparent between SDHS

154! ! ! ! ! ! ! and polygyny across all Egernia species investigated. Low sample sizes precluded statistical testing of the hypothesis of male-biased SDHS being associated with polygyny.

Within the Egernia clade, differences in size have been attributed to intrasexual selection due to male–male competition (e.g. Bull and Pamula 1996;

Osterwalder et al. 2004; Chapple, 2005). Male aggression is documented in several

Egernia group species (Chapple, 2003; Clemann et al., 2004; While et al., 2009;

Michael et al., 2010), and several factors may contribute to this, including competition for limited shelter sites, and resource-defense polygyny. In line with this, L. kintorei is a burrowing species that constructs extensive tunnel systems in which close kin live, and offspring mature (McAlpin et al. 2011). These burrow systems represent a significant energy investment into a stable shelter site that is shared with mates and offspring, and it is likely that males will defend such a resource, and exclude other males within their territories who attempt to court their females. Larger heads would likely be of benefit in male–male encounters, perhaps even allowing other males to assess rivals before such encounters escalate to combat.

The elusive nature, underground lifestyle and remote habitat of Liopholis kintorei make these lizards extremely difficult to observe in the wild. However, although not yet observed in L. kintorei, male–male aggression and formation of male dominance hierarchies have been documented in other species within the Egernia group (Arena and Wooller, 2003; Chapple, 2003; Osterwalder et al., 2004).

For some lizards, differences in relative head size between the sexes is suggested to be a result of differential allometry of the sexes (Cox et al., 2007;

Gifford and Powell, 2007). Upon reaching maturity, females may direct more energy to the growth of morphometric characters pertaining to reproduction. For example, higher fecundity may be attained by rapidly increasing SVL to increase potential

! 155! ! ! ! ! ! ! clutch or litter size, perhaps at the expense of traits less-directly related to female reproductive output such as head size (Cooper and Vitt, 1989). This may play a role in divergence of relative head size between the sexes in L. kintorei, a viviparous lizard whose reproductive output may be increased in females of a relatively larger body size.

Level of territoriality does not predict extent of SDHS in lizards

Contrary to expectations, the model showed no significant relationship between the extent of SDHS and territoriality in lizards. There are several possible explanations for this.

While many territorial species do exhibit SDHS, and this may still be related to intrasexual competition in those species, other selective forces may be driving dimorphism in less territorial species to the same end. For example, as mentioned earlier, females may allocate more energy to reproductive growth (e.g. body size) at the expense of other traits such as head size (Cooper and Vitt, 1989). Perhaps, selection for male head size, per se, is not always the source of head size dimorphism, but differential allometry of the sexes plays a role. Ecological niche separation may also contribute to SDHS in the absence of sexual selection (Shine,

1989). Numerous squamate species show evidence of intersexual differences in trophic morphology and bite force consistent with dietary partitioning (Cox et al.,

2007; Herrel et al., 1996), which is related to dimorphism in head dimensions and reduces intersexual competition for food resources. However, numerous studies have not been able to rule out sexual selection and attribute SDHS solely to niche partitioning (Herrel et al., 1996, 1999; Vincent and Herrel, 2007). It is quite possible that differences in diet are a consequence, rather than a cause, of SDHS (Vincent and Herrel, 2007).

! 156! ! ! ! ! ! ! Both the degree of SDHS and territoriality can be influenced by numerous factors relating to seasonality, habitat and demography. For example, in the skink,

Plestiodon laticeps, male head size is under androgenic control and actually increases during the mating season (Vitt and Cooper, 1985; Vitt and Cooper, 1986).

In many species, males tend to be more aggressive toward conspecifics during the mating season, and many of these may even form nonagonistic aggregations at other times of the year (Martins, 1994). Female biased sex ratios and high female densities have been known to trigger male territorial behaviour in typically non- territorial species (Stamps, 1983; Stamps et al., 1997). Measurements and quantification of territoriality made by different researchers at different times may result in inter-observer error and result in misclassification of some ‘non-territorial’ species as ‘territorial’ and vice-versa. Habitat structure can also impact upon territorial behaviour. Butler et al. (2000) found that differences in territorial behaviour

(and SSD) in Greater Antillean Anoles, such that those living in more open habitat with higher visibility, adopted sit-and-wait foraging strategies and tended to be more territorial, while species living in more complex habitats with less visibility tended towards active foraging and lower territoriality (also discussed in Stamps, 1977).

Some error may lie in the choice of territoriality as a measure of intrasexual selection. Defining species as belonging to one of three categories of territoriality may be over simplistic. Territorial behaviour can manifest through a continuum of behaviours, from scent marking territory to patrolling territories and aggressive combat (Maher and Lott, 1995). Martins (1994) suggests ten categories of territoriality that incorporate defense ‘style’ (e.g. combat, threat, avoidance) as well as defense area (as in Stamps, 1977), however there was not sufficient detail in the literature to enable us to use this system to categorise many of the species in this study. Cox et al. (2003) found stronger correlations between SSD in body size and

! 157! ! ! ! ! ! ! continuous measures of intrasexual selection (e.g. male–female home range ratio) and suggested that categorical measures may conceal the variation in intensity of selection needed to elucidate associations with SSD.

Additionally, perhaps a male’s ability to win contests, rather than territoriality per se is of primary importance in explaining sexual dimorphism in head size. For example, whether an individual defends a home range or a specific site, it would still benefit from larger head size in the event of an agonistic encounter. That is, the selective pressure for SDHS may be the same irrespective of the area defended. In the case of some non-territorial or widely ranging species, aggression is still known to occur when males encounter one another (Marco and Perez-Mellado, 1999; Gvozdik and Van Damme, 2003). For example, mate-guarding in non-territorial or monogamous species is known (Marco and Perez-Mellado, 1999; Bull, 2000), and males of these species may still be under selection pertaining to male–male combat.

Furthermore, much of the literature may be biased towards those species whose behaviour is more conspicuous and easily observed, as is the case in many of the

Agamidae (Stamps, 1977; Doody et al., 2013). Lack of observation of territorial behaviour may lead to some territorial species being classified as non-territorial.

The next step for this study would be to incorporate continuous measures of intrasexual selection into the predictive model of SDHS, such as home range ratios between the sexes and female home range size (Stamps, 1983; Cox et al., 2003), to provide more comprehensive information on the interplay between head size, intrasexual competition and the proximate causes for dimorphism. Future research on the relationship between reproductive success and SDHS would benefit from more studies that directly measure male reproductive success, such as genetic paternity analyses (e.g. Lebas, 2001), as well as controlled experimental studies. For example, manipulation of factors such as territory size, habitat quality, food

! 158! ! ! ! ! ! ! availability, female density or sex ratios and how these translate into reproductive success might help to elucidate the potential ecological, physiological and behavioural mechanisms involved in intrasexual selection and the evolution of SSD, and may also shed light on the behaviours of more elusive and non-territorial species.

Acknowledgements

Thank you to T. Hibbitts, D. Noble, J. Riley, A. Stow and M. Whiting for supplying unpublished data for this analysis, to D. Nipperess, D. Noble and H. Baldwin for statistical advice, and K. Umbers for helpful comments on the manuscript. Sampling of Liopholis kintorei would not have been possible without help from numerous field assistants. Thank you to the Australian Wildlife Conservancy for permission to sample at their Newhaven property, in particular D. Moore and J. Schofield whose logistical support and local knowledge on site were invaluable.

Literature cited

Andersson, M., and R. A. Norberg. 1981. Evolution of reversed sexual size

dimorphism and role partitioning among predatory birds, with a size scaling of

flight performance. Biological Journal of the Linnean Society 15:105–130.

Arena, P. C., and R. D. Wooller. 2003. The reproduction and diet of Egernia kingii

(Reptilia: Scincidae) on Penguin Island, Western Australia. Australian Journal of

Zoology 51:495–504.

Bates, D. 2005. Fitting linear mixed models in R. Using the lme4 package. R News

5:27–30.

! 159! ! ! ! ! ! ! Bateson, Z. W., Krenz, J. D., and R. E. Sorensen. 2011. Multiple paternity in the

common five-lined skink (Plestiodon fasciatus). Journal of Herpetology 45:504–

510.

Beck, D. D., and R. D. Jennings. 2003. Habitat use by Gila monsters: the importance

of shelters. Herpetological Monographs 17:111–129.

Becker, B., and M. Paulissen. 2012. Sexual dimorphism in head size in the little

brown skink (Scincella lateralis). Herpetological Conservation and Biology

7:109–114.

Bisazza, A. 1993. Male competition, female mate choice and sexual size dimorphism

in poeciliid fishes. Marine Behaviour and Physiology 23:257–286.

Borja, M. 1985. Spatial and temporal behaviour of Gallotia galloti in a natural

population of Tenerife. Bonn Zoological Bulletin 36:541–552.

Braña, F. 1996. Sexual dimorphism in lacertid Lizards: male head increase vs female

abdomen increase? Oikos 75:511–523.

Bull, C. M. 2000. Monogamy in lizards. Behavioural Processes 51:7–20.

Bull, C. M., and Y. Pamula. 1996. Sexually dimorphic head sizes and reproductive

success in the sleepy lizard Tiliqua rugosa. Journal of Zoology 240:511–521.

Butler, M. A., Schoener, T. W., and J. B. Losos. 2000. The relationship between

sexual size dimorphism and habitat use in Greater Antillean Anolis lizards.

Evolution 54:259–272.

Cabrera, M. P., Scrocchi, G. J., and F. B. Cruz. 2013. Sexual size dimorphism and

allometry in Liolaemus of the L. laurenti group (Sauria: Liolaemidae):

morphologic lability in a clade of lizards with different reproductive modes.

Zoologischer Anzeiger 252:299–306.

! 160! ! ! ! ! ! ! Carazo, P., Font, E., and R. Desfilis. 2011. The role of scent marks in female choice

of territories and refuges in a lizard (Podarcis hispanica). Journal of Comparative

Psychology 125:362–365.

Carazo, P., Noble, D. W., Chandrasoma, D., and M. J. Whiting. 2014. Sex and

boldness explain individual differences in spatial learning in a lizard.

Proceedings of the Royal Society B: Biological Sciences 281:20133275.

Carothers, J. 1984. Sexual selection and sexual dimorphism in some herbivorous

lizards. The American Naturalist 124:244–254.

Castilla, A. M., and D. Bauwens. 1990. Reproductive and fat body cycles of the

lizard, Lacerta lepida, in central Spain. Journal of Herpetology 24:261–266.

Censky, E. J. 1995. Mating strategy and reproductive success in the teiid lizard,

Ameiva plei. Behaviour 132:529–557.

Chapple, D. G. 2003. Ecology, life-history, and behavior in the Australian Scincid

genus Egernia, with comments on the evolution of complex sociality in lizards.

Herpetological Monographs 17:145–180.

Chapple, D. G. 2005. Life history and reproductive ecology of White’s skink, Egernia

whitii. Australian Journal of Zoology 53:353–360.

Chapple, D., and J. Keogh. 2005. Complex mating system and dispersal patterns in a

social lizard, Egernia whitii. Molecular Ecology 14:1215–1227.

Clemann, N., Chapple, D. G., and J. Wainer. 2004. Sexual dimorphism, diet, and

reproduction in the swamp skink, Egernia coventryi. Journal of Herpetology

38:461–467.

Cooper, W. E., and L. J. Vitt. 1987. Deferred agonistic behavior in a long-lived scincid

lizard Eumeces laticeps. Oecologia 72:321–326.

Cooper, W. E., and L. J. Vitt. 1989. Sexual dimorphism of head and body size in an

iguanid lizard: paradoxical results. The American Naturalist 133:729–735.

! 161! ! ! ! ! ! ! Cordes, I., Mouton, P., and J. van Wyk. 1995. Sexual dimorphism in two girdled

lizard species, Cordylus niger and Cordylus cordylus. South African Journal of

Zoology 30:187–196.

Cox, R. M., Skelly, S. L., and H. B. John-Alder. 2003. A comparative test of adaptive

hypotheses for sexual size dimorphism in lizards. Evolution 57:1653–1669.

Cox, R. M., Butler, M. A., and H. B. John-Alder. 2007. The evolution of sexual size

dimorphism in reptiles. In: Fairbairn, D. J., Blanckenhorn, W. U., Szekely, T.,

(Editors). Sex, size and gender roles: evolutionary studies of sexual size

dimorphism. Oxford University Press, Oxford, UK.

Doody, J. S., Burghardt, G. M., and V. Dinets. 2013. Breaking the social–non-social

dichotomy: a role for reptiles in vertebrate social behavior research? Ethology

119:1–9.

Doughty, P., Maryan, B., Melville, J., and J. Austin. 2007. A new species of

Ctenophorus (Lacertilia: Agamidae) From Lake Disappointment, Western

Australia. Herpetologica 63:72–86.

Du, Y., Lin, C.-X., Lin, L.-H., Qiu, Q.-B., and X. Ji. 2011. Ontogenetic shifts in sexual

dimorphism and female reproduction in the Reeves’s butterfly lizard Leiolepis

reevesii from Hainan, China. Journal of Herpetology 45:399–405.

Dubey, S., Chevalley, M., and R. Shine. 2011. Sexual dimorphism and sexual

selection in a montane scincid lizard (Eulamprus leuraensis). Austral Ecology

36:68–75.

Duckett, P. E., Morgan, M. H., and A. J. Stow. 2012. Tree-dwelling populations of the

skink Egernia striolata aggregate in groups of close kin. Copeia 2012:130–134.

Edwards, J. R., and S. P. Lailvaux. 2013. Do interspecific interactions between

females drive shifts in habitat use? A test using the lizards Anolis carolinensis

and A. sagrei. Biological Journal of the Linnean Society 110:843–851.

! 162! ! ! ! ! ! ! El Mouden, E. H., Znari, M., and R. P. Brown. 1999. Skeletochronology and mark-

recapture assessments of growth in the North African agamid lizard (Agama

impalearis). Journal of Zoology 249:455–461.

Fairbairn, J., and R. Shine. 1993. Patterns of sexual size dimorphism in seabirds of

the southern hemisphere. Oikos 68:139–145.

Fenner, A. L., Pavey, C. R. & Bull, C. M. 2012. Behavioural observations and use of

burrow systems by an endangered Australian arid-zone lizard, Slater's skink

(Liopholis slateri). Australian Journal of Zoology 60:127–132.

Font, E., Barbosa, D., Sampedro, C., and P. Carazo. 2012. Social behavior, chemical

communication, and adult neurogenesis: studies of scent mark function in

Podarcis wall lizards. General and Comparative Endocrinology 177:9–17.

Galán, P. 2000. Females that imitate males: dorsal coloration varies with

reproductive stage in female Podarcis bocagei (Lacertidae). Copeia 2000:819–

825.

Galán, P. 2008. Ontogenetic and sexual variation in the coloration of the lacertid

lizards Iberolacerta monticola and Podarcis bocagei. Do the females prefer the

greener males? Animal Biology 58:173–198.

Gardner, M. G., Bull, C. M., Cooper, S. J., and G. A. Duffield. 2001. Genetic evidence

for a family structure in stable social aggregations of the Australian lizard

Egernia stokesii. Molecular Ecology 10:175–183.

Gienger, C. M., and D. D. Beck. 2007. Heads or tails? Sexual dimorphism in

helodermatid lizards. Canadian Journal of Zoology 85:92–98.

Gifford, M. E., and R. Powell. 2007. Sexual dimorphism and reproductive

characteristics in five species of Leiocephalus lizards from the Dominican

Republic. Journal of Herpetology 41:521–527.

! 163! ! ! ! ! ! ! Gittleman, J. L., and M. Kot. 1990. Adaptation: statistics and a null model for

estimating phylogenetic effects. Systematic Zoology 39:227–241.

Griffith, H. 1991. Heterochrony and evolution of sexual dimorphism in the fasciatus

group of the scincid genus Eumeces. Journal of Herpetology 25:24–30.

Gvozdik, L., and R. van Damme. 2003. Evolutionary maintenance of sexual

dimorphism in head size in the lizard Zootoca vivipara: a test of two hypotheses.

Journal of Zoology 259:7–13.

Harlow, P. S., and J. E. Taylor. 2000. Reproductive ecology of the jacky dragon

(Amphibolurus muricatus): an agamid lizard with temperature-dependent sex

determination. Austral Ecology 25:640–652.

Healey, M., Uller, T., and M. Olsson. 2007. Seeing red: morph-specific contest

success and survival rates in a colour-polymorphic agamid lizard. Animal

Behaviour 74:337–341.

Hernández-Salinas, U., Ramírez-Bautista, A., Leyte-Manrique, A., and G. R. Smith.

2010. Reproduction and sexual dimorphism in two populations of Sceloporus

grammicus (Sauria: Phrynosomatidae) from Hidalgo, Mexico. Herpetologica

66:12–22.

Herrel, A., van Damme, R., and F. de Vree. 1996. Sexual dimorphism of head size in

Podarcis hispanica atrata: testing the dietary divergence hypothesis by bite force

analysis. Netherlands Journal of Zoology 46:253–262.

Herrel, A., Spithoven, L., van Damme, R., and F. de Vree. 1999. Sexual dimorphism

of head size in Gallotia galloti: testing the niche divergence hypothesis by

functional analyses. Functional Ecology 13:289–297.

Herrel, A., McBrayer, L., and P. Larson. 2007. Functional basis for sexual differences

in bite force in the lizard Anolis carolinensis. Biological Journal of the Linnean

Society 91:111–119.

! 164! ! ! ! ! ! ! Hews, D. K. 1990. Examining hypotheses generated by field measures of sexual

selection on male lizards, Uta palmeri. Evolution 44:1956–1966.

Hews, D. K. 1993. Food resources affect female distribution and male mating

opportunities in the iguanian lizard Uta palmeri. Animal Behaviour 46:279–291.

Hews, D. K. 1996. Size and scaling of sexually-selected traits in the lizard, Uta

palmeri. Journal of Zoology 238:743–757.

Huang, W. 1998. Sexual size dimorphism and microhabitat use of two sympatric

lizards, Sphenomorphus taiwanensis and Takydromus hsuehshanensis, from the

Central highlands of Taiwan. Zoological Studies 37:302–308.

Huang, W. 2006. Ecology and reproductive patterns of the grass lizard, Takydromus

sauteri, in a tropical rain forest of an east Asian island. Journal of Herpetology

40:267–273.

Jenssen, T. A., Marcellini, D. L., and P. H. Goforth. 1989. Differential infanticide by

adult curly-tailed lizards, Leiocephalus schreibersi. Animal Behaviour 38:1054–

1061.

Ji, X., Qiu, Q.-B., and C. H. Diong. 2002. Sexual dimorphism and female reproductive

characteristics in the oriental garden lizard, Calotes versicolor, from Hainan,

southern China. Journal of Herpetology 36:1–8.

Kikukawa, N., and T. Hikida. 2012. Geometric morphometric analysis of sexual

dimorphism and allometric growth in the skull of the Japanese grass lizard,

Takydromus tachydromoides (Reptilia: Lacertidae). Current Herpetotogy 31:97–

106.

Kohlsdorf, T., Garland, T., and C. A. Navas. 2001. Limb and tail lengths in relation to

substrate usage in Tropidurus lizards. Journal of Morphology 248:151–164.

! 165! ! ! ! ! ! ! Kwiatkowski, M. A., and B. K. Sullivan. 2002. Mating system structure and population

density in a polygynous lizard, Sauromalus obesus (= ater). Behavioral Ecology

13:201–208.

Lappin, A., Hamilton, P., and B. Sullivan. 2006. Bite-force performance and head

shape in a sexually dimorphic crevice-dwelling lizard, the common chuckwalla

[Sauromalus ater (= obesus)]. Biological Journal of the Linnean Society 88:215–

222.

Lebas, N. R. 2001. Microsatellite determination of male reproductive success in a

natural population of the territorial ornate dragon, Ctenophorus ornatus.

Molecular Ecology 10:193–203.

López, P., and J. Martín. 2001. Fighting rules and rival recognition reduce costs of

agression in male lizards, Podarcis hispanica. Behavioral Ecology and

Sociobiology 49:111–116.

Lovich, J. E., and J. W. Gibbons. 1992. A review of techniques for quantifying sexual

size dimorphism. Growth, Development and Aging 56:269–281.

Luo, L., Wu, Y., Zhang, Z., and X. Xu. 2012. Sexual size dimorphism and female

reproduction in the white-striped grass lizard Takydromus wolteri. Current

Zoology 58:236–243.

Madsen, T. 1983. Growth rates, maturation and sexual size dimorphism in a

population of grass snakes, Natrix natrix, in southern Sweden. Oikos 40:277–

282.

Maher, C. R., and D. F. Lott. 1995. Definitions of territoriality used in the study of

variation in vertebrate spacing systems. Animal Behaviour 49:1581–1597.

Marco, A., and V. Perez-Mellado. 1999. Mate guarding, intrasexual competition and

mating success in males of the non-territorial lizard Lacerta schreiberi. Ethology

Ecology and Evolution 11:279–286.

! 166! ! ! ! ! ! ! Martins, E. 1994. Phylogenetic perspectives on the evolution of lizard territoriality. In:

Vitt, L. J. and Pianka, E., (Editors). Lizard Ecology: Historical and Experimental

Perspectives. Princeton University Press, Princeton, New Jersey.

Martins, E., Labra, A., Halloy, M., and J. Thompson. 2004. Large-scale patterns of

signal evolution: an interspecific study of Liolaemus lizard head bob displays.

Animal Behaviour 68:453–463.

McAlpin, S., Duckett, P., and A. Stow. 2011. Lizards cooperatively tunnel to construct

a long-term home for family members. PLOS ONE 6:e19041.

McGuire, J. A., and R. Dudley. 2011. The biology of gliding in flying lizards (genus

Draco) and their fossil and extant analogs. Integrative and Comparative Biology

51:983–990.

Michael, D. R., Cunningham, R. B., and D. B. Lindenmayer. 2010. The social elite:

habitat heterogeneity, complexity and quality in granite inselbergs influence

patterns of aggregation in Egernia striolata (Lygosominae: Scincidae). Austral

Ecology 35:862–870.

Mouton, P., and J. van Wyk. 1993. Sexual size dimorphism in cordylid lizards: a case

study of the Drakensberg crag lizard, Pseudocordylus melanotus. Canadian

Journal of Zoology 71:1715–1723.

Mouton, P., Flemming, A., and E. Kanga. 1999. Grouping behaviour, tail-biting

behaviour and sexual dimorphism in the armadillo lizard (Cordylus cataphractus)

from South Africa. Journal of Zoology 249:1–10.

O’Connor, D., and R. Shine. 2003. Lizards in “nuclear families”: a novel reptilian

social system in Egernia saxatilis (Scincidae). Molecular Ecology 12:743–752.

Olsson M., and T. Madsen. 1998. Sexual selection and sperm competition in reptiles.

In:Sperm Competition and Sexual Selection (eds Birkhead TR, Møller AP). pp.

504–575, Academic Press, London.

! 167! ! ! ! ! ! ! Olsson, M., Healey, M., Wapstra, E., Schwartz, T., Lebas, N., and T. Uller. 2007.

Mating system variation and morph fluctuations in a polymorphic lizard.

Molecular Ecology 16:5307–5315.

Osterwalder, K., Klingenbock, A., and R. Shine. 2004. Field studies on a social lizard:

home range and social organisation in an Australian skink, Egernia major.

Austral Ecology 29:241–249.

Paulissen, M. A., Meyer, H. A., and T. S. Hibbs. 2013. Movement patterns and

sociality of the Mediterranean Gecko, Hemidactylus turcicus, in southwestern

Louisiana. The Southwestern Naturalist 58:344–350.

Perry, G. 1996. The evolution of sexual dimorphism in the lizard Anolis polylepis

(Iguania): evidence from intraspecific variation in foraging behavior and diet.

Canadian Journal of Zoology 74:1238–1245.

Pinto, A., Wiederhecker, H., and G. Colli. 2005. Sexual dimorphism in the Neotropical

lizard, Tropidurus torquatus (Squamata, Tropiduridae). Amphibia-Reptilia

26:127–137.

Pyron, R., Burbrink, F., and J. Wiens. 2013. A phylogeny and revised classification of

Squamata, including 4161 species of lizards and snakes. BMC Evolutionary

Biology 13:93–145.

Qi, Y., Noble, D., Fu, J., and M. Whiting. 2012. Spatial and social organization in a

burrow-dwelling lizard (Phrynocephalus vlangalii) from China. PLOS ONE

7:e41130.

Radder, R., Shanbhag, B., and S. Saidapur. 2001. Ontogeny of sexual size

dimorphism in the tropical garden lizard, Calotes versicolor (Daud.). Journal of

Herpetology 35:156–160.

Randriamahazo, H. 2000. Sexual size dimorphism in the lizard Oplurus cuvieri cuvieri

(Squamata, Opluridae) from Madagascar. African Zoology 35:287–293.

! 168! ! ! ! ! ! ! Reaney, L. T., and M. J. Whiting. 2002. Life on a limb: ecology of the tree agama

(Acanthocercus a. atricollis) in southern Africa. Journal of Zoology 257:439–448.

Reaney, L. T., and M. J. Whiting. 2003. Are female tree agamas (Acanthocercus

atricocollis atricollis) turned on my males or resources? Ethology Ecology and

Evolution 15:19–30.

Ribeiro, L. B., Sousa, B. M., and S. C. Gomides. 2009. Range structure, microhabitat

use, and activity patterns of the saxicolous lizard Tropidurus torquatus

(Tropiduridae) on a rock outcrop in Minas Gerais, Brazil. Revista Chilena de

Historia Natural 82:577–588.

Ribeiro, L. B., Gogliath, M., de Sales, R. F., and E. M. Freire. 2011. Mating behavior

and female accompaniment in the whiptail lizard Cnemidophorus ocellifer

(Squamata, Teiidae) in the Caatinga region of northeastern Brazil. Biota

Neotropica 11:363–368.

Robles, C. I., and M. Halloy. 2010. Core area overlap in a neotropical lizards,

Liolaemus quilmes: relationship with territoriality and reproductive strategy.

Herpetological Journal 20:243–248.

Saenz, D., and R. Conner. 1996. Sexual dimorphism in head size of the

Mediterranean gecko Hemidactylus turcicus (Sauria: Gekkonidae). Texas

Journal of Science 48:207–212.

Schoener, T. W., and A. Schoener. 1982. Intraspecific variation in home-range size

in some Anolis lizards. Ecology 63:809–823.

Shine, R. 1978. Sexual size dimorphism and male combat in snakes. Oecologia

33:269–277.

Shine, R. 1979. Sexual selection and sexual dimorphism in the Amphibia. Copeia

1979:297–306.

! 169! ! ! ! ! ! ! Shine, R. 1989. Ecological causes for the evolution of sexual dimorphism: a review of

the evidence. The Quarterly Review of Biology 64:419–461.

Shine, R. 1994. Sexual size dimorphism in snakes revisited. Copeia 1994:326–346.

Shine, R., Keogh, S., Doughty, P., and H. Giragossyan. 1998. Costs of reproduction

and the evolution of sexual dimorphism in a “flying lizard” Draco melanopogon

(Agamidae). Journal of Zoology 246:203–213.

Sinn, D. L., While, G. M., and E. Wapstra. 2008. Maternal care in a social lizard: links

between female aggression and offspring fitness. Animal Behaviour 76:1249–

1257.

Smith, R. J. 1999. Statistics of sexual size dimorphism. Journal of Human Evolution

36:423–459.

Stamps, J. 1977. Social behaviour and spacing patterns in lizards. In: Gans, C. and

Tinkle, D. (Editors). Biology of Reptilia. Academic Press, New York.

Stamps, J. 1983. Sexual selection, sexual dimorphism and territoriality. In: Huey, R.

B., Pianka, E. R., and Schoener, T. W. (Editors). Lizard Ecology: Studies of a

Model Organism. Harvard University Press, Cambridge, Massachusetts, USA.

Stamps, J. A., Losos, J. B., and R. M. Andrews. 1997. A comparative study of

population density and sexual size dimorphism in lizards. American Naturalist

149:65–90.

Stow, A. J., and P. Sunnucks. 2004. High mate and site fidelity in Cunningham’s

skinks (Egernia cunninghami) in natural and fragmented habitat. Molecular

Ecology 13:419–430.

Stow, A. J., Sunnucks, P., Briscoe, D. A., and M. G. Gardner. 2001. The impact of

habitat fragmentation on dispersal of Cunningham’s skink (Egernia

cunninghami): evidence from allelic and genotypic analyses of microsatellites.

Molecular Ecology 10:867–878.

! 170! ! ! ! ! ! ! Uller, T., and M. Olsson. 2008. Multiple paternity in reptiles: patterns and processes.

Molecular Ecology 17:2566–2580.

Van der Meer, M. H., Whiting, M. J., and W. R. Branch. 2010. Ecology of southern

African Sandveld lizards (Lacertidae, Nucras). Copeia 2010:568–577.

Van Sluys, M., Rocha, C. F., Vrcibradic, D., Galdino, C. A., and A. F. Fontes. 2004.

Diet, activity, and microhabitat use of two syntopic Tropidurus species (Lacertilia:

Tropiduridae) in Minas Gerais, Brazil. Journal of Herpetology 38:606–611.

Vincent, S. E., and A. Herrel. 2007. Functional and ecological correlates of

ecologically-based dimorphisms in squamate reptiles. Integrative and

Comparative Biology 47:172–188.

Vitt, L. J. 1983. Reproduction and sexual dimorphism in the tropical teiid lizard

Cnemidophorus ocellifer. Copeia 1983:359–366.

Vitt, L. J. 1986. Reproductive tactics of sympatric gekkonid lizards with a comment on

the evolutionary and ecological consequences of invariant clutch size. Copeia

1986:773–786.

Vitt, L. J. 1991a. Ecology and life history of the scansorial arboreal lizard Plica plica

(Igunanidae) in Amazonian Brazil. Canadian Journal of Zoology 69:504–511.

Vitt, L. J. 1991b. Ecology and life-history of the wide-foraging lizard Kentropyx

calcarata (Teiidae) in Amazonian Brazil. Canadian Journal of Zoology 69:2791–

2799.

Vitt, L. J., and W. E. Cooper. 1985. The evolution of sexual dimorphism in the skink

Eumeces laticeps: an example of sexual selection. Canadian Journal of Zoology

63:995–1002.

Vitt, L. J., and W. E. Cooper. 1986. Skink reproduction and sexual dimorphism:

Eumeces fasciatus in the Southeastern United States, with notes on Eumeces

inexpectatus. Journal of Herpetology 20:65–76.

! 171! ! ! ! ! ! ! Vitt, L. J., and C. de Carvalho. 1992. Life in the trees: the ecology and life history of

Kentropyx strintus (Teiidae) in the lavrado area of Roraima, Brazil, with

comments on the life histories of tropical teiid lizards. Canadian Journal of

Zoology 70:1995–2006.

Vitt, L. J., and P. A. Zani. 1996. Ecology of the elusive tropical lizard Tropidurus

[=Uracentron] flaviceps (Tropiduridae) in lowland rain forest of Ecuador.

Herpetologica 52:121–132.

Vitt, L., Zani, P., Caldwell, J., and E. Carrillo. 1995. Ecology of the lizard Kentropyx

pelviceps (Sauria: Teiidae) in lowland rain forest of Ecuador. Canadian Journal

of Zoology 73:691–703.

Weckerly, F. W. 1998. Sexual-size dimorphism: influence of mass and mating

systems in the most dimorphic mammals. Journal of Mammalogy 79:33–52.

While, G. M., Uller, T., and E. Wapstra. 2009a. Family conflict and the evolution of

sociality in reptiles. Behavioral Ecology 20:245–250.

While, G. M., Uller, T., and E. Wapstra. 2009b. Within-population variation in social

strategies characterise the social and mating system of an Australian lizard,

Egernia whitii. Austral Ecology 34:938–949.

While, G. M., Uller, T. & Wapstra, E. (2011) Variation in social organization

influences the opportunity for sexual selection in a social lizard. Molecular

Ecology 20:844–852.

Whiting, M. J., Reaney, L. T., and J. S. Keogh. 2007. Ecology of Wahlberg’s velvet

gecko, Homopholis wahlbergii, in southern Africa. African Zoology 42:38–44.

Znari, M., and E. H. El Mouden. 1997. Sexual dimorphism, reproductive and fat body

cycles in Bibron’s agama (Agama impalearis, Boettger, 1874) (Sauria:

Agamidae). Herpetologica 53:411–422.

!

! 172! ! ! ! ! ! ! Supplementary material 5.1

Table S5.1. Pearson correlation coefficients for head width, head length and head height in Liopholis kintorei. Values for females (n = 51) are above the diagonal, and males (n = 29) below. All values are significant (P < 0.05).

Head width! Head length! Head height! ! Head width! ,! 0.909! 0.795!

Head length! 0.937! ,! 0.759!

Head height! 0.842! 0.779! ,!

!

! 173! ! ! ! ! ! ! ! !

! 174! ! ! ! ! ! ! ! CHAPTER 6

General discussion

Photo by Martin Whiting

! 175! ! ! ! ! ! ! Overview

Over the past decade, studies on lizards of the Egernia group have provided extensive insight into the early evolution of social living. The diversity of social organisation and social behaviours within the group, including basic parental care, have provided a valuable opportunity to compare potential fundamental drivers of the convergent evolution of these behaviours across taxonomically diverse groups.

Because of the interest surrounding the ‘social’ Egernia species, research has largely focused on several key species that exhibit high levels of social organisation (e.g.

Egernia cunninghami, Stow and Sunnucks, 2004; E. stokesii, Duffield and Bull, 2002;

E. saxatilis, O’Connor and Shine, 2003; Liopholis whitii, Chapple and Keogh, 2005).

Each of these live in close association with rocky outcrops where they shelter in crevices. In order to identify the environmental or behavioral features that lead to social organisation and behavior, it is valuable to obtain information on species in the

Egernia group that exhibit different behaviours or life history, including those that live in different habitats (e.g. Egernia coventryi, Clemann et al., 2004), or are more solitary (e.g. Liopholis inornata, Daniel, 1998).

Unlike other crevice dwelling, group-living Egernia group species, L. kintorei may not be as limited by shelter site availability when it can construct its own, although the availability of suitable habitat features may yet be a limiting factor

(Chapter 4). With its burrowing lifestyle, and potentially relatively high levels of investment into offspring (McAlpin et al, 2011), the great desert skink (Liopholis kintorei) promises to provide valuable insight into the relative importance of ecological constraints and kin-selection in the maintenance of sociality with further studies.

Information on the life history and habits of L. kintorei is extremely difficult to gather for a number of reasons: many of the areas in which they are known to occur

! 176! ! ! ! ! ! ! are extremely remote and difficult to access, and finding new populations is difficult for the same reasons; their occupancy of underground burrow systems prohibits direct observation of many social interactions among group members; and, in the case of this study, their threatened status meant that permission could not be obtained to carry out manipulative or captive experiments, making it difficult to directly test hypotheses surrounding social interactions and parental care (discussed below in Suggestions for future research). Therefore, the primary aim of my research was to use molecular tools to investigate patterns of genetic structure at fine (e.g. within burrows) and broad (e.g. across their distribution) scales, and combine this with recapture methods to shed light on aspects of the species’ biology, including movement patterns and social organisation. The outcomes of this research are summarised below, and their implications discussed in a broader evolutionary and conservation context, with suggestions for future research.

1. Liopholis kintorei is patchily distributed across a vast geographic range extending approximately 1.3 million km2. Despite the lack of substantial geographic barriers in the Australian arid zone, habitat specialisation exhibited by many desert species may limit connectivity among populations. This was investigated in L. kintorei, whose habitat specialisation and limited dispersal was predicted to limit gene flow among localities where they are found. Samples were collected from six localities spread across their distribution, and mitochondrial ND4 sequence data and microsatellite genotypic data were generated. I found substantial differentiation between three main regions, the extent of which suggests they should be managed separately, in particular the southeastern locality close to Uluru.

2. By combining spatial genetic analysis with mark-recapture data, I demonstrated male-biased dispersal and fine-scale movement in L. kintorei.

Juveniles and adult females showed high natal-site fidelity, inferred by high and

! 177! ! ! ! ! ! ! significant pairwise relatedness at short distances, while there was no genetic structure among males. Complementary to this, recapture data revealed that males travelled a greater distance between burrow systems, and used a higher number of distinct burrow systems than females.

3. Social organisation, including group composition, relatedness and stability was investigated using molecular and mark-recapture data collected over three years

(2011–2013). Group composition varied from adult pairs to a mated pair with three cohorts of offspring. There was a high level of variation in relatedness and stability among groups; however, relatedness was generally higher between group members than between individuals from different groups. More than half of the groups identified had at least two members present across two or more years, although the genetic data suggests that the stability of some of these groups is even higher

(minimum four years). I demonstrated that active burrow systems are excavated in close proximity to termite mounds, which represent a valuable and stable food resource in the unpredictable desert environment.

4. I explored the hypothesis that sexual selection is responsible for sexual dimorphism in head size (SDHS) in lizards. The Egernia group exhibits a range of mating systems, from monogamy through to high levels of polygyny, and so I began by investigating the relationship between polygyny and the degree of SDHS within this monophyletic group. Although no trend was evident, the amount of published data available was insufficient to statistically test this relationship. I therefore extended my investigation to encompass lizards from a range of taxa. Little data are available on polygyny in lizards, and so the level of territoriality was used as a proxy.

I found that SDHS is generally male-biased in lizards, however I found no difference in the degree of SDHS between species categorised as non-territorial, specific-site defense territorial, and home-range defense territorial.

! 178! ! ! ! ! ! !

Evolutionary implications

The evolution of kin-based sociality has long been of considerable interest to researchers of evolutionary and behavioural biology, and identifying the mechanisms and processes involved in its origin and maintenance continues to generate much discourse and even controversy to this day (Nowak et al., 2010; Okasha, 2010;

Strassmann et al., 2011). Kin selection and ecological constraints are invoked as key drivers for the evolution of sociality (Hamilton, 1964; Emlen, 1982). In particular, behaviours associated with kin-based social living such as parental care are thought to arise when environmental conditions constrain dispersal, and then high competition for resources increase the relative benefits of tolerance of family members within ones territory (Klug et al., 2012). More recently, there has been much discussion on the benefits of philopatry and prolonged association with parents: the benefits of group living, rather than the costs and benefits of delayed dispersal per se (Stacey and Ligon, 1991; Komdeur, 1992). This may, in turn, enable the development of more complex social bonds and cooperative behaviours.

The presence of highly related groups sheltering together in L. kintorei

(Chapter 4) is in line with patterns observed in other Egernia group species (Gardner et al., 2001; O’Connor and Shine, 2003; Stow and Sunnucks, 2004a; Chapple and

Keogh, 2005). The relative ubiquity of family living within the Egernia group despite its rarity among other reptiles (Chapple, 2003) suggests that the behaviour is monophyletic (Gardner et al., 2008), and that it is phylogenetically conserved within the group. The high variability in group composition, relatedness and stability of L. kintorei groups (Chapter 4), could be a function of differences in their preferred habitat, and reduced constraints associated with available shelter sites given their burrowing rather than saxicolous life history. However, the construction of a burrow

! 179! ! ! ! ! ! ! system requires a significant investment of time and energy, and simply building a new shelter site when required may not be so easy. Further, a newly constructed burrow might not be equivalent to those burrows that have been established over multiple generations. McAlpin (2001) found that most newly established L. kintorei burrows fail, probably because small, simple burrows offer less protection than the large complex tunnel systems characteristic of the older, more established burrow complexes.

The finding that active burrow systems were positively associated with termite mounds (e.g. a stable food resource) suggests that ecological constraints may contribute to the maintenance of group living in L. kintorei (Chapter 4). Group living in the cordylid lizard, Ouroborus cataphractus, is thought to have been driven by the lizards’ dependence on termites as a food source (Shuttleworth et al., 2008; Mouton,

2011). Similarly, the presence of termites as an abundant and concentrated food source may be important for the persistence of L. kintorei groups, where competition for food within the family group may otherwise be too high in the unpredictable desert environment to maintain group living.

The sympatric presence of numerous species, including the closely related L. inornata, a solitary burrowing lizard (Daniel, 1998, Chapple et al., 2004), in the southern portion of L. kintorei’s distribution suggests that ecological constraints imposed by the desert environment may not be the only factors facilitating kin-based social living. Given that sociality appears to be ancestral to the Egernia group

(Gardner et al., 2008), it is most likely that L. inornata has lost this trait, while it has been maintained in L. kintorei. This could be due to less constrained habitat requirements of L. inornata, which constructs small burrows in sandy or loamy soils, however is found in a range of habitat types, with a much broader distribution spanning both arid and more temperate climates (Daniel, 1998).

! 180! ! ! ! ! ! ! Competition among kin should drive dispersal from the natal territory unless the costs associated with it are balanced by the direct and/or inclusive benefits of remaining with relatives (Perrin and Mazalov, 2000; Davis Rabosky et al., 2012). Social interactions among kin may also facilitate group cohesion. For example, juvenile desert night lizards (Xantusia vigilis) are more likely to aggregate and less likely to disperse when released with relatives (Davis, 2011), and in Liopholis whitii, adult males only tolerate their own offspring in their territories, and the corresponding risk of infanticide might prevent females from mating with extra-group males (While et al.,

2009). The burrow systems in which L. kintorei live represent a significant investment of energy (McAlpin et al., 2011), and adult individuals may increase their inclusive fitness by sharing their burrow with kin to the exclusion of unrelated individuals. It is likely that the delayed dispersal of juvenile L. kintorei into adulthood (Chapter 3) has both direct and indirect benefits, such as access to resources within their parents’ territory (O’Connor and Shine, 2004), or protection from predators and aggressive conspecifics (Lanham and Bull, 2004; Sinn et al., 2008; While et al., 2009). In addition to increased survival, and in turn, higher inclusive fitness within the group, natal philopatry might also reinforce social bonds among relatives: a precursor to kin- based sociality.

The diversity of mating systems within the Egernia group provides not only an opportunity to investigate social organisation in reptiles, but understanding the processes responsible for the diversity of social systems is also essential to our understanding of major biological processes including sexual selection (Chapter 5) and dispersal (Chapter 3).

! 181! ! ! ! ! ! ! Conservation implications

Liopholis kintorei is a threatened species, nationally listed as Vulnerable (IUCN,

2014), and considered Endangered in South Australia (National Parks and Wildlife

Act, 1972). Several targeted monitoring and management actions have been undertaken across South Australia, Western Australia, and the Northern Territory

(Chapter 2), however, little is known about the movement patterns, as well as broad- and fine-scale genetic structure of this species: information that would be important when designing management practices such as fire management or translocations.

Despite the general lack of geographic barriers throughout the Australian arid zone, Chapter 2 uncovered considerable genetic divergence among regions containing L. kintorei, indicating restricted dispersal throughout their range. This, combined with the isolation of localities implied by their patchy occurrence in the landscape (McAlpin, 2001; McAlpin et al., 2011), suggests that migration is unlikely to replenish genetic diversity if it is reduced over time. Erosion of genetic diversity and viability of localities containing L. kintorei is a real threat given the small census

(McAlpin, 2001) and genetic Ne (Chapter 2) estimates of population size, which could, in turn, render localised parts of their distribution vulnerable to inbreeding and inbreeding depression (Stow and Sunnucks, 2004b). Translocations have been demonstrated to reverse the effects of inbreeding depression (Madsen et al., 1996;

Weeks et al., 2011), however if this need eventuated, areas chosen for translocations would need to be carefully selected to avoid the potential of outbreeding depression.

Three main regions were identified for management: Uluru in the southeast,

Newhaven and Sangster’s Bore to the north, and Watarru, Docker River and

Warburton to the southwest. Further sampling, particularly in the southwest of their distribution, is required to confirm the extent of differentiation, although Uluru was found to be sufficiently distinct that it should be considered separately for

! 182! ! ! ! ! ! ! management. Much of the distribution of L. kintorei is remote and virtually inaccessible, making efforts to extensively map, sample and monitor L. kintorei populations logistically difficult and costly. At the very least, conservation management in areas containing L. kintorei should prioritise the preservation of suitable habitat and address localised changes in fire regimes and predation pressure to reduce the risk of further population decline (see Moore et al., 2014).

At a finer scale, sex-biased dispersal and relatively low-vagility evidenced by stability of groups and burrow occupancy across years (Chapters 3 and 4) also have important implications for management practices. Since European settlement, higher fuel loads as a result of fire suppression have resulted in large-scale wildfire events that severely reduce the suitable habitat of many organisms. Historically, patch burning was used as a tool by indigenous peoples to hunt and gather, and this created a mosaic of successional stages of vegetation at a fine scale that promoted broader-scale biodiversity (Morton et al., 2011). Such practices are used now by land managers to mitigate the threat of wildfires, and have been suggested as a management tool for conservation of L. kintorei habitat (McAlpin, 2001). Moore et al.

(2014) showed that after a clean burn (i.e. all vegetation removed, as occurs during wildfire), L. kintorei burrows had severely reduced occupancy, and this was less extreme when patch burning was implemented. The authors suggested that reduced vegetation cover after a clean burn resulted in increased exposure and higher risk of predation. If male L. kintorei are more mobile than females among burrows (Chapter

3), they may be more vulnerable to predation than females. In areas affected by fire, sex-based differences in their propensity to disperse could result in low recolonisation of these areas because of low dispersal among females (Chapter 3), and low vagility of established groups (Chapter 4). If viable populations do not occur nearby, recolonisation, if it occurs at all, would be considerably delayed. This should

! 183! ! ! ! ! ! ! be taken into account for management, and monitoring of natural recolonisation of areas post-fire should be undertaken, assisting if required. If higher vagility does indeed make males more vulnerable to predation, fire management practices should aim to maintain suitable habitat by preserving ground cover to provide protection from predators.

Suggestions for future research

The Egernia group is an ideal model by which to examine the origin and maintenance of group living in vertebrates, given the inter- and intraspecific diversity in mating systems and social behaviour that it displays, and a well-resolved phylogeny onto which these traits can be mapped (Chapple, 2003; Gardner et al. 2008). In particular, some of the arid-zone members of the Liopholis genus could provide a unique comparative system for further study. Liopholis kintorei, L. striata and L. inornata are obligately burrowing species that construct burrows in sandy soils (Chapple, 2003), and do not appear to be constrained by habitat availability when they can construct their own tunnels (rather than utilising rocky crevices as in other Egernia group species). These three species have overlapping distributions and occur sympatrically in some areas (Chapple et al., 2004; S. McAlpin, pers. comm.), and yet show different levels of social organisation: L. kintorei lives in groups of highly related individuals (Chaper 3), L. inornata is solitary (Daniel, 1998), and anecdotal evidence has suggested that L. striata lives in pairs, sometimes with its offspring (Henzell,

1972; Pianka and Giles, 1982). Investigating potential drivers of social organisation in sympatric populations of these species may shed light on processes driving the maintenance of social living (in L. kintorei), its secondary loss (in L. inornata) or transitions between the two (L. striata).

! 184! ! ! ! ! ! ! With regard to extending the current study of L. kintorei, long-term monitoring of the study population at Newhaven will help us to identify important ecological drivers promoting group cohesion and dynamism. The arid zone of Australia experiences unpredictable rainfall events, with temporal variability higher than most other deserts at a given mean annual rainfall (Morton et al., 2011). Further, longer- term temporal variation is contributed to by climatic oscillations with return times of between four years (El Niño Southern Oscillation) and 19 years (Pacific Oscillation;

Morton et al., 2011). These oscillations lead to ‘resource pulses’ in the desert environment, and can greatly alter the availability of food resources, which in turn influences breeding patterns and population densities of desert animals (Letnic and

Dickman, 2010). If this food availability is a real limitation for L. kintorei, we might see variation in group living coincident with these climatic fluctuations and associated resource availability.

Valuable information could be obtained from monitoring group structure and social organisation across a temporal scale broader than the three years of this study, and investigating the relationship with ecological and demographic variables such as food resource availability and predator population sizes.

Given the elusive nature and underground lifestyle of L. kintorei, it is extremely difficult to monitor grouping behaviours and associations within the burrows directly.

Therefore, our observations of social organisation were restricted to animals trapped at burrow systems overnight. In order to maximise our encounter rate with the lizards, we deployed remote PIT-tag readers at numerous burrows in the latter two years of the study. Despite the fact that we were limited in the number of PIT-tag readers relative to Elliott traps (in 2012, there was a total of ~3800 trap nights compared to

330 PIT-read nights), the readers contributed substantially to the overall number of lizards recorded in their burrows. Thirty-one percent of our records of lizards at their

! 185! ! ! ! ! ! ! burrows in 2012 came from PIT-tag records rather than Elliott trapping, despite the limited number of readers we had access to (8 in total). Because L. kintorei burrows at our site had up to 17 entrances, we did not have enough readers for sufficient coverage of the tunnel entrances to rely heavily on these for data collection.

However, this method is less intrusive than trapping, and if more readers could be acquired for future studies, they would be invaluable for more intensive monitoring of movement in and out of burrows.

The potentially high level of parental care and cooperation in L. kintorei warrants further investigation. McAlpin et al. (2011) found that multiple group members contribute to maintenance of burrow systems, and suggested that parents invest energy into such a home that provides protection and a direct survival benefit to offspring. Parental care is often much harder to assess in lizards than in other taxa, due in part to the chemosensory, rather than visual or auditory nature of much of lizard communication (Doody et al., 2013). Previous work in Egernia group species has indirectly addressed parental care by showing that the offspring of more aggressive females have increased survival (Sinn et al., 2008), and that within-group offspring benefit from access to their parents’ territories and associated resources

(O’Connor and Shine, 2004; While et al., 2009). Despite some evidence of parental care (Masters and Shine, 2003; Sinn et al., 2008; While et al., 2009), the extent of active defense of offspring as opposed to defense of territory is uncertain. This could be delineated by manipulative experiments such as aggression trials of females in the presence of their offspring. Furthermore, detailed investigation of the burrowing behaviour of L. kintorei groups, and the relative investment of individuals from different age and sex classes would provide valuable insight into the cooperative behaviours and sociality of this species, and shed light on parental care: Do males invest less in burrow construction when extra-pair offspring are resident? Do female

! 186! ! ! ! ! ! ! group members forego reproduction and cooperatively contribute to burrow maintenance?

Cross-fostering studies can also provide valuable insight into studies of kin- selection in social lizards. Davis (2011) used a cross-fostering approach to manipulate relatedness between ‘social’ mothers and siblings. She found that juveniles had lower propensity to disperse in the presence of their biological mothers than did fostered juveniles; and when released into the field, juveniles were more likely to aggregate when released with biological rather than foster siblings. Similar studies in L. kintorei would help to tease apart whether sociality is a result of positive kin interaction rather than constraints on dispersal, and could be used to assess cooperative behaviours in the presence of kin vs. non-kin, and the relative importance of direct and inclusive fitness benefits.

In Chapter 4, I found a positive relationship between the presence of active burrow systems and the presence of termite mounds at my study site, and suggested that this may facilitate group cohesion by providing an abundant and temporally stable food source. Although beyond the scope of my work in this thesis, it would be of value to extend this analysis across a broader geographic and temporal range.

Determining whether group size and stability, as well as the longevity of burrows are related to presence of termites within suitable habitat for L. kintorei would also be of value in assessing the importance of these stable food resources for group living. If there were a strong relationship between termite presence and L. kintorei, this information would also be of benefit to conservation managers making decisions for fire management and preservation of habitat, or for use in models to map potential L. kintorei habitat.

Recent advances in genetic statistical modeling are also providing important information on gene flow and dispersal capabilities in a range of species (Jombart et

! 187! ! ! ! ! ! ! al., 2008; McRae et al., 2008; Dudaniec et al., 2013). The ability to link landscape and genetic data from various temporal and spatial scales allows analyses to be carried out at resolutions relevant to broad ecological processes, or at the scale of management units (Dudaniec et al., 2013). Dudaniec et al. (2013) developed a model to analyse landscape–genetic relationships that incorporates resistance modeling to assess patterns of dispersal through varied and dynamic environments. Such analyses would be useful to L. kintorei management that could incorporate extensive data available on geology, vegetation, fire history, and climate. This would allow managers to assess the relative importance of these factors in facilitating or impeding gene flow among habitat patches. In turn, this could help land managers to improve connectivity within and among management areas.

Conclusion

Despite the growing literature on social behaviour in reptiles, many questions remain unanswered. Many lizards are cryptic in their behaviours (Doody et al., 2013), which, combined with the relatively new addition of reptiles to the literature on sociality, means that more detailed information on social behaviours and interactions among kin is required across a broader range of reptiles. By revealing new details about spatial and temporal genetic patterns and social living, this study provides new insights into the biology and social organisation of L. kintorei, an elusive and unique species. This thesis also provides a foundation upon which further behavioural studies can be pursued to extend our understanding of factors driving sociality in this species. Further, there is now information on dispersal patterns and genetic connectivity to inform future conservation practices of this threatened species in central arid Australia.

! 188! ! ! ! ! ! ! Literature cited

Chapple, D. G. 2003. Ecology, life-history, and behavior in the Australian Scincid

genus Egernia, with comments on the evolution of complex sociality in lizards.

Herpetological Monographs 17:145–180.

Chapple, D. G., and J. S. Keogh. 2005. Complex mating system and dispersal

patterns in a social lizard, Egernia whitii. Molecular Ecology 14:1215–1227.

Chapple, D. G., Keogh, J. S., and M. N. Hutchinson. 2004. Molecular

phylogeography and systematics of the arid-zone members of the Egernia whitii

(Lacertilia: Scincidae) species group. Molecular Phylogenetics and Evolution

33:549–561.

Clemann, N., Chapple, D. G., and J. Wainer. 2004. Sexual dimorphism, diet, and

reproduction in the swamp skink, Egernia coventryi. Journal of Herpetology

38:461–467.

Daniel, M. C. 1998. Aspects of the ecology of Rosen’s desert skink, Egernia inornata,

in the Middleback Ranges, Eyre Peninsula. Department of Environmental

Science and Management. University of Adelaide, South Australia.

Davis, A. 2011. Kin presence drives philopatry and social aggregation in juvenile

Desert Night Lizards (Xantusia vigilis). Behavioral Ecology 23:18–24.

Davis Rabosky, A., Corl, A., Liwanag, H., Surget-Groba, Y., and B. Sinervo. 2012.

Direct fitness correlates and thermal consequences of facultative aggregation in

a desert lizard. PLOS ONE 7:e40866.

Doody, J. S., Burghardt, G. M., and V. Dinets. 2013. Breaking the social – non-social

dichotomy!: a role for reptiles in vertebrate social behavior research? Ethology

119:1–9.

Dudaniec, R. Y., Rhodes, J. R., Worthington Wilmer, J., Lyons, M., Lee, K. E.,

McAlpine, C. A., and F. N. Carrick. 2013. Using multilevel models to identify

! 189! ! ! ! ! ! ! drivers of landscape-genetic structure among management areas. Molecular

Ecology 22:3752–65.

Duffield, G. A., and C. M. Bull. 2002. Stable social aggregations in an Australian

lizard, Egernia stokesii. Naturwissenschaften 89:424–427.

Emlen, S. T. 1982. The evolution of helping I. An ecological constraints model. The

American Naturalist 119:29–39.

Gardner, M. G., Bull, C. M., Cooper, S. J., and G. A. Duffield. 2001. Genetic evidence

for a family structure in stable social aggregations of the Australian lizard

Egernia stokesii. Molecular Ecology 10:175–183.

Gardner, M. G., Hugall, A. F., Donnellan, S. C., Hutchinson, M. N., and R. Foster.

2008. Molecular systematics of social skinks: phylogeny and taxonomy of the

Egernia group (Reptilia: Scincidae). Zoological Journal of the Linnean Society

154:781–794.

Hamilton, W. 1964. The Genetical Evolution of Social Behaviour I. Journal of

Theoretical Biology 7:1–16.

Henzell, R. P. 1972. Adaptation to aridity in lizards of the Egernia whitii species-

group. PhD Thesis, University of Adelaide, South Australia.

IUCN. (2014). IUCN Red List of Threatened Species. Version 2014.2. Retrieved from

www.iucnredlist.org.

Jombart, T., Devillard, S., Dufour, A.-B., and D. Pontier. 2008. Revealing cryptic

spatial patterns in genetic variability by a new multivariate method. Heredity

101:92–103.

Klug, H., Alonzo, S., and M. Bonsall. 2012. Theoretical foundations of parental care.

In: Royle, N., Smiseth, P., and Kolliker, M. (Editors). The Evolution of Parental

Care (1st Edition). Oxford University Press, Oxford, UK.

! 190! ! ! ! ! ! ! Komdeur, J. 1992. Importance of habitat saturation and territory quality for evolution

of cooperative breeding in the Seychelles warbler. Nature 358:493–495.

Lanham, E. J., and C. M. Bull. 2004. Enhanced vigilance in groups in Egernia

stokesii, a lizard with stable social aggregations. Journal of Zoology, 263:95–99.

Letnic, M., and C. R. Dickman. 2010. Resource pulses and mammalian dynamics:

conceptual models for hummock grasslands and other Australian desert

habitats. Biological Reviews 85:501–521.

Madsen, T., Stille, B., and R. Shine. 1996. Inbreeding depression in an isolated

population of adders Vipera berus. Biological Conservation 75:113–118.

Masters, C., and R. Shine. 2003. Sociality in lizards: Family structure in free-living

King’s skinks Egernia kingii from southwestern Australia. Australian Zoologist

32:377–380.

McAlpin, S. 2001. A recovery plan for the great desert skink (Egernia kintorei) 2001-

2011. Arid Lands Environment Centre, Alice Springs, Australia.

McAlpin, S., Duckett, P., and A. Stow. 2011. Lizards cooperatively tunnel to construct

a long-term home for family members. PLOS ONE 6:e19041.

McRae, B. H., Dickson, B. G., Keitt, T. H., and V. B. Shah. 2008. Using circuit theory

to model connectivity in ecology, evolution, and conservation. Ecology 89:2712–

2724.

Moore, D., Kearney, M., Paltridge, R., McAlpin, S., and A. Stow. 2014. Is fire a

threatening process for Liopholis kintorei, a nationally listed threatened skink?

Wildlife Research, in press.

Morton, S. R., Stafford Smith, D. M., Dickman, C. R., Dunkerley, D. L., Friedel, M. H.,

McAllister, R. R., Reid, J. R., Roshier, D. A., Smith, M. A., Walsh, F. J., Wardle,

G. M., Watson, I. W., and M. Westoby. 2011. A fresh framework for the ecology

of arid Australia. Journal of Arid Environments 75:313–329.

! 191! ! ! ! ! ! ! Mouton, P. L. 2011. Aggregation behaviour of lizards in the arid western regions of

South Africa. African Journal of Herpetology 60:155–170.

Nowak, M. A., Tarnita, C. E., and E. O. Wilson. 2010. The evolution of eusociality.

Nature 466:1057–1062.

O’Connor, D. E., and R. Shine. 2003. Lizards in “nuclear families”: a novel reptilian

social system in Egernia saxatilis (Scincidae). Molecular Ecology 12:743–752.

O’Connor, D. E., and R. Shine. 2004. Parental care protects against infanticide in the

lizard Egernia saxatilis (Scincidae). Animal Behaviour 68:1361–1369.

Okasha, S. 2010. Altruism researchers must cooperate. Nature 467:653–655.

Perrin, N., and V. Mazalov. 2000. Local competition, inbreeding, and the evolution of

sex-biased dispersal. The American Naturalist 155:116–127.

Pianka, E. R., and W. F. Giles. 1982. Notes on the ecology of two species of

nocturnal skinks, Egernia inornata and Egernia striata, in the Great Victoria

Desert. The West Australian Naturalist 15:8-13.

Shuttleworth, C., Mouton, P. F., and J. H. van Wyk. 2008. Group size and termite

consumption in the armadillo lizard, Cordylus cataphractus. Amphibia-Reptilia

29:171–176.

Sinn, D. L., While, G. M., and E. Wapstra. 2008. Maternal care in a social lizard: links

between female aggression and offspring fitness. Animal Behaviour 76:1249–

1257.

Stacey, P., and J. Ligon. 1991. The benefits-of-philopatry hypothesis for the evolution

of cooperative breeding: variation in territory quality and group size effects. The

American Naturalist 137:831-846.

Stow, A. J., and P. Sunnucks. 2004a. High mate and site fidelity in Cunningham’s skinks (Egernia cunninghami) in natural and fragmented habitat. Molecular Ecology

13:419–430.

! 192! ! ! ! ! ! ! Stow, A. J., and P. Sunnucks. 2004b. Inbreeding avoidance in Cunningham’s skinks

(Egernia cunninghami) in natural and fragmented habitat. Molecular Ecology

13:443–447.

Strassmann, J. E., Page, R. E., Robinson, G. E., and T. D. Seeley. 2011. Kin

selection and eusociality. Nature 471: E5-E6.

Weeks, A. R., Sgro, C. M., Young, A. G., Frankham, R., Mitchell, N. J., Miller, K. A.,

Byrne, M., Coates, D. J., Eldridge M. D., Sunnucks, P., Breed, M., James, E. A.,

and A. A. Hoffmann. 2011. Assessing the benefits and risks of translocations in

changing environments: a genetic perspective. Evolutionary Applications 4:709–

725.

While, G. M., Uller, T., and E. Wapstra. 2009. Family conflict and the evolution of

sociality in reptiles. Behavioral Ecology 20:245–250.

! 193! ! ! ! ! ! !

!

! 194! ! ! ! ! ! ! APPENDIX

Photo by Martin Whiting

! 195! ! ! ! ! ! ! Table A1. Morphometric data for all Liopholis kintorei individuals captured at the main site in 2011

head head head capture sample age weight svl sex width length height date id class (g) (mm) (mm) (mm) (mm) 23/10/11 L0101 F juv 108 21.62 29.32 19.99 154 24/10/11 L0102 unknown juv 100 21.01 29.58 19.61 154 20/10/11 L0201 M juv 111 21.52 30.52 20.28 157 22/10/11 L0202 M juv 124 22.15 29.82 21.51 158 24/10/11 L0203 unknown juv 110 22.04 29.23 19.35 158 26/10/11 L0401 M adult 186 26.19 34.15 22.24 185 15/10/11 L0501 F juv 90 20.61 28.25 18.71 148 21/10/11 L0502 F juv 87 20.08 27.6 18.17 145 21/10/11 L0503 F juv 83 20.61 27.3 18.3 145 23/10/11 L0504 F juv 80 20.46 27.6 17.82 146 23/10/11 L0505 unknown juv 20 18.58 19.75 12.03 93 29/10/11 L0506 unknown juv 24 14.47 20.33 12.72 98 19/10/11 L0601 F adult 169 25.2 33.23 23.3 175 24/10/11 L0701 F juv 118 22.06 29.92 19.88 153 25/10/11 L0702 M adult 174 25.19 34.14 22.81 175 18/10/11 L0801 F adult 190 25.65 35.22 22 181 23/10/11 L0802 unknown juv 96 20.26 27.53 20.59 147 12/10/11 L0901 M adult 242 30.33 39.08 26.04 199 24/10/11 L0902 F adult 196 26.55 33.41 23.31 186 10/10/11 L1001 unknown juv 102 22.07 29.2 20.84 152 10/10/11 L1301 unknown juv 106 20.49 27.99 19.14 151 16/10/11 L1302 unknown juv 23 14.18 20.45 12.63 100 20/10/11 L1303 M juv 25 14.16 20.18 12.63 98 21/10/11 L1304 M adult 241 30.63 37.53 25.81 190 26/10/11 L1305 F adult 230 28.19 35.78 25.41 196 22/10/11 L1401 unknown juv 81 20.55 26.7 16.88 145 25/10/11 L1402 F juv 108 21.81 28.16 19.8 154 26/10/11 L1403 unknown juv 90 21.25 27.8 20.44 144 18/10/11 L1601 M adult 296 30.7 40.03 26.87 209 23/10/11 L1602 M adult 201 27.62 37.47 25.38 187 20/10/11 L1701 M adult 165 24.11 32.2 22.35 172 26/10/11 L1702 F adult 126 23.13 30.69 20.18 160 23/10/11 L1901 unknown juv 98 20.87 29.86 20.71 151 18/10/11 L2401 F juv 95 21.03 29.08 19.07 154 20/10/11 L2402 M adult 143 23.6 30.83 22.05 164 20/10/11 L2403 unknown juv 105 21.55 28.63 19.72 156 22/10/11 L2404 F juv 99 21.18 28.57 19.27 151 22/10/11 L2405 F juv 89 20.5 27.83 17.6 142 26/10/11 L2406 F adult 228 29.21 36.99 26.63 191

! 196! ! ! ! ! ! ! Table A2. Morphometric data for all Liopholis kintorei individuals captured at the main site in 2012

(NB. two sample periods in 2012)

OCTOBER SAMPLING

head head head capture sample age weight svl sex width length height date id class (g) (mm) (mm) (mm) (mm) 8/10/12 L0101 F adult 122 23.89 31.04 23.11 172 8/10/12 L0103 F adult 204 29.12 35.18 24.89 200 9/10/12 L0201 M adult 116 23.44 31.67 23.13 173 15/10/12 L0202 M adult 142 25.40 32.65 22.46 176 15/10/12 L0501 F adult 108 24.04 30.28 19.40 160 16/10/12 L0502 F adult 92 22.56 29.75 20.23 163 14/10/12 L0503 F adult 96 23.07 28.61 21.22 163 30/10/12 L0504 F adult 122 23.37 30.19 20.92 165 15/10/12 L0505 unknown juv 30 16.59 22.24 15.05 115 31/10/12 L0507 F juv 64 19.10 25.92 16.91 132 3/11/12 L0601 F adult 164 27.14 33.18 23.10 184 8/10/12 L0701 F adult 130 23.91 30.99 22.50 165 15/10/12 L0702 M adult 160 27.24 34.21 23.35 180 9/10/12 L0802 unknown juv 102 22.57 29.25 21.14 157 9/10/12 L0803 M adult 198 29.03 36.05 26.66 196 31/10/12 L0902 F adult 206 27.14 32.72 25.53 197 27/10/12 L1101 F adult 188 27.90 34.34 23.82 189 12/10/12 L1303 M juv 40 16.91 23.17 16.25 115 8/10/12 L1304 M adult 216 30.20 37.63 30.20 197 9/10/12 L1305 F adult 198 28.18 33.30 26.51 196 27/10/12 L1402 F adult 148 24.11 30.05 20.46 165 9/10/12 L1601 M adult 254 30.76 40.07 28.11 211 8/10/12 L1602 M adult 202 28.27 35.13 25.85 193 13/10/12 L1701 M adult 178 27.04 33.82 26.82 183 9/10/12 L1702 F adult 142 24.95 32.06 23.00 177 31/10/12 L1703 F adult 198 28.77 34.19 25.45 198 3/11/12 L1902 F adult 236 28.33 34.33 24.62 191 20/10/12 L2301 M adult 134 24.96 31.62 22.61 175 3/11/12 L2401 F adult 126 24.43 31.19 22.30 175 8/10/12 L2402 M adult 143 26.08 32.51 23.03 172 11/10/12 L2404 F adult 126 23.56 30.71 21.05 170 13/10/12 L2405 F juv 92 22.69 29.02 22.10 153

! 197! ! ! ! ! ! !

head head head capture sample age weight svl sex width length height date id class (g) (mm) (mm) (mm) (mm) 12/10/12 L2406 F adult 214 29.37 35.93 28.45 195 12/10/12 L2407 unknown juv 14 13.34 17.95 12.20 86 3/11/12 L3601 M adult 140 25.74 31.69 22.94 170

31/10/12 L4301 F adult 132 24.07 30.28 21.58 171

DECEMBER SAMPLING

head head head capture sample age weight svl sex width length height date id class (g) (mm) (mm) (mm) (mm) 14/12/12 L0101 F adult 142 25.02 31.58 22.97 173 16/12/12 L0103 F adult 222 28.08 34.99 27.90 200 23/12/12 L0201 M adult 152 25.30 33.85 22.95 180 14/12/12 L0202 M adult 170 27.20 33.76 25.40 177 27/12/12 L0204 F adult 236 27.93 33.74 26.33 197 16/12/12 L0501 F adult 134 25.15 31.34 22.50 171 16/12/12 L0502 F adult 130 24.46 30.79 21.63 167 16/12/12 L0503 F adult 120 24.06 29.56 21.66 168 16/12/12 L0505 unknown juv 46 18.06 23.69 16.50 123 16/12/12 L0507 F juv 82 20.84 27.00 18.86 145 16/12/12 L0508 F adult 240 28.26 34.45 24.41 196 31/12/12 L0509 unknown juv 10 11.60 16.28 11.03 75 16/12/12 L0601 F adult 184 27.35 33.93 25.94 187 27/12/12 L0701 F adult 154 24.90 32.13 23.21 172 14/12/12 L0803 M adult 224 30.17 36.76 27.67 196 14/12/12 L1303 M juv 60 19.73 26.18 17.26 133 14/12/12 L1304 M adult 220 30.76 37.35 26.18 198 23/12/12 L1402 F adult 144 24.85 28.78 21.68 175 23/12/12 L1501 M adult 174 26.39 33.17 23.82 184 14/12/12 L1601 M adult 262 31.24 40.70 31.47 211 14/12/12 L1602 M adult 196 28.71 36.22 26.23 194 14/12/12 L1701 M adult 192 27.55 34.41 27.37 187 14/12/12 L1801 F adult 152 26.20 32.92 23.58 182 23/12/12 L1902 F adult 236 27.87 33.43 26.08 192 16/12/12 L2301 M adult 154 26.17 32.69 23.31 184 16/12/12 L2405 F adult 130 24.25 30.46 22.11 167 14/12/12 L2407 unknown juv 30 15.94 20.85 12.73 105

! 198! ! ! ! ! ! !

head head head capture sample age weight svl sex width length height date id class (g) (mm) (mm) (mm) (mm) 30/12/12 L3601 M adult 174 27.18 33.53 24.82 180 16/12/12 L4301 F adult 138 24.99 31.42 23.68 175 14/12/12 L4302 M adult 240 30.83 37.52 29.11 195 27/12/12 L4501 unknown juv 12 12.21 16.26 9.85 78 31/12/12 L4502 F adult 176 27.39 34.25 24.73 189 8/01/13 L4503 unknown juv 10 12.35 17.37 10.92 81 30/12/12 L5201 M adult 162 26.66 33.73 25.15 183

! 199! ! ! ! ! ! ! Table A3. Morphometric data for all Liopholis kintorei individuals captured at sites other than the main site 2012

head head head capture sample age weight svl sex width length height date id class (g) (mm) (mm) (mm) (mm) 29/10/11 C20101 F adult 210 26.43 33.19 25.51 183 28/10/11 C20301 M adult 120 23.07 30.21 20.49 163 28/10/11 C20302 F adult 252 27.09 36.22 24.22 195 29/10/11 C20303 unknown juv 126 22.64 30.68 20.24 159 28/10/11 C20801 unknown juv 46 16.74 22.91 12.3 114 29/10/11 C20802 unknown juv 50 16.92 23.28 16.35 121 29/10/11 C20803 F adult 192 26.35 32.84 23 174 2/12/12 C3801 M adult 148 23.84 31.10 20.84 169 3/12/12 C3802 unknown juv 34 15.40 20.52 13.82 109 5/12/12 C4501 F adult 244 28.00 33.02 26.68 189 5/12/12 C5101 F adult 262 29.61 35.26 26.51 204 2/12/12 C10701 M adult 254 29.45 36.75 29.23 196 2/12/12 C10702 unknown juv 22 14.48 19.62 13.28 97 5/12/12 C11601 unknown juv 60 18.44 23.88 17.93 136 28/11/12 C20201 M adult 252 30.31 37.50 26.16 193 25/11/12 C20301 M adult 180 25.20 33.89 25.16 186 28/11/12 C20302 F adult 250 27.38 35.03 24.18 198 29/11/12 C20701 F adult 252 28.76 34.17 25.48 188 29/11/12 C20702 M adult 268 30.28 36.48 30.21 197 25/11/12 C20801 unknown juv 110 22.78 29.47 21.31 159 28/11/12 C21103 unknown juv 26 14.88 20.12 13.46 100 25/11/12 C22201 M adult 202 27.65 33.95 25.39 189 28/11/12 C22202 F adult 154 25.23 31.90 21.81 182 26/11/12 C22401 M adult 244 30.41 37.59 26.38 210 28/11/12 C22402 F adult 244 28.46 34.04 24.97 189 29/11/12 C22501 F adult 148 24.76 30.73 22.48 172 28/11/12 C22701 F adult 228 28.00 33.11 24.94 186 26/11/12 C22801 F adult 240 27.64 33.77 24.05 193 26/11/12 C22901 F adult 232 29.05 35.16 23.68 198 10/11/12 A0301 unknown juv 28 14.27 20.12 12.84 101 10/11/12 A0302 M juv 114 21.86 29.65 20.94 158 12/11/12 A0303 unknown juv 66 18.22 25.18 17.08 138 11/11/12 A0701 F adult 212 27.74 34.22 26.19 189 23/11/12 A0901 F adult 232 28.35 36.28 27.29 198 18/11/12 LTR0101 F adult 134 23.98 30.91 21.85 171 23/11/12 LTR0102 F adult 172 25.70 32.30 21.31 177 15/11/12 LTR1201 unknown adult 150 26.58 33.60 26.37 175 23/11/12 LTR1301 M adult 230 30.15 37.15 31.17 198 18/11/12 LTR1401 F adult 132 23.96 29.53 21.30 161 20/11/12 HH0101 M adult 236 29.08 36.55 25.33 200 6/12/12 NY0901 unknown juv 114 23.02 29.41 20.69 155 8/12/12 NY0902 unknown juv 90 21.62 27.27 18.53 147

! 200! ! ! ! ! ! ! head head head capture sample age weight svl sex width length height date id class (g) (mm) (mm) (mm) (mm) 7/12/12 NY1501 M adult 210 30.65 36.25 26.61 201 22/11/12 LB0901 unknown juv 96 21.56 28.25 20.36 148 28/11/12 LB0902 unknown juv 76 19.80 26.33 18.79 137 22/11/12 LB1201 F adult 282 29.85 34.50 25.51 194 18/11/12 CK0201 M adult 236 30.17 35.46 29.38 199 11/11/12 CK0801 F adult 204 29.48 37.26 26.75 196 18/11/12 CK0802 unknown juv 76 20.92 27.45 17.91 148 18/11/12 CK0803 F adult 198 27.45 33.85 23.43 189 18/11/12 CK1001 F adult 156 24.89 31.28 22.43 177 15/11/12 CK1401 F adult 208 28.47 34.60 26.36 188 21/11/12 CK1402 F adult 238 27.79 33.92 23.85 194 23/11/12 CK1501 M adult 216 28.74 35.27 28.22 199

! 201! ! ! ! ! ! ! Table A4. Morphometric data for all Liopholis kintorei individuals captured at the main site in 2013

head head head capture sample age weight svl sex width length height date id class (g) (mm) (mm) (mm) (mm) 11/10/13 L0201 M adult 164 27.58 35.45 27.98 185 9/10/13 L0505 unknown juv 80 21.21 27.08 19.35 147 9/10/13 L0507 F adult 112 21.58 30.01 23.80 165 9/10/13 L0508 F adult 198 27.96 34.23 27.91 197 28/10/13 L0509 unknown juv 20 14.29 19.18 12.45 98 31/10/13 L0510 unknown juv 22 14.74 20.22 12.51 103 9/10/13 L0601 F adult 188 27.33 34.21 24.55 187 9/10/13 L0902 F adult 200 27.67 32.87 25.30 192 13/10/13 L1303 M adult 100 23.86 30.69 20.13 160 13/10/13 L1306 unknown juv 32 16.91 22.34 15.37 118 7/10/13 L1402 F adult 159 25.65 31.27 22.50 179 7/10/13 L1501 M adult 190 28.46 34.40 24.50 190 17/10/13 L1603 unknown juv 26 15.89 21.52 14.24 109 9/10/13 L1701 M adult 186 29.18 33.75 26.64 192 11/10/13 L1703 F adult 218 29.17 33.03 27.71 197 1/11/13 L1704 unknown juv 26 15.44 20.12 14.06 108 13/10/13 L1902 F adult 238 28.17 34.11 27.44 192 13/10/13 L2301 M adult 182 28.68 34.49 27.31 191 7/10/13 L2401 F adult 166 26.87 33.22 23.45 186 9/10/13 L2405 F adult 162 25.81 31.80 25.87 175 7/10/13 L2406 F adult 226 29.42 35.53 25.92 196 7/10/13 L2407 unknown juv 60 19.67 25.66 17.58 136 21/10/13 L3601 M adult 174 28.13 34.79 25.62 187 11/10/13 L4301 F adult 160 25.64 31.51 22.13 182 24/10/13 L4302 M adult 216 31.56 37.76 25.20 201 28/10/13 L4303 F adult 194 28.44 35.21 24.51 191 13/10/13 L4401 F adult 182 26.82 32.13 23.23 186 9/10/13 L4501 unknown juv 30 15.81 20.99 13.49 111 30/10/13 L4502 F adult 202 27.16 34.34 23.99 191 9/10/13 L4503 unknown juv 22 15.07 20.32 14.32 103 13/10/13 L5201 M adult 140 28.08 34.48 26.06 184

! 202! ! ! ! ! ! ! Table A5. GPS waypoints (WGS84 UTMs) for all burrows at which lizards were captured throughout the study. Only the main site (Honeymoon Lake) was trapped every year. All other sites were only trapped in 2012

Site Burrow ID Eastings Northings HONEYMOON LAKE L01 728568 7482184 (MAIN SITE) L02 728516 7482017 L04 728344 7482000

L05 728610 7481954

L06 728613 7482022

L07 728599 7482098

L08 728597 7482178

L09 728565 7481864

L10 728553 7481750

L11 728452 7481920

L12 728438 7481956

L13 728440 7482089

L14 728472 7482167

L15 728476 7482200

L16 728487 7482102

L17 728394 7481992

L18 728445 7482089

L19 728415 7482052

L20 728586 7481977

L23 728539 7482095

L24 728521 7482158

L28 728500 7482113

L32 728582 7481916

L33 728472 7482113

L36 728520 7482188

L39 728376 7482119

L43 728355 7482047

L44 728448 7482031

L45 728519 7481927

L48 728405 7482073

L50 728419 7482147

L52 728354 7482022

CAMEL BORE RD 1 C38 723670 7477700 C45 723648 7477769

C51 723925 7477819

C78 724197 7476819

C106 724225 7476363

C107 723588 7477739

C116 723747 7477738

C202 724158 7479673

C203 724139 7479623 C207 724306 7479616

! 203! ! ! ! ! ! !

Site Burrow ID Eastings Northings CAMEL BORE RD 1 C208 724317 7479658 (cont.) C211 724029 7479675 C222 723846 7479777

C224 723844 7479818

C225 723868 7479849

C227 724000 7479729

C228 724061 7479690

C229 723922 7479751

CUCKOO RD CK02 719068 7477981 CK08 719520 7477813

CK10 719156 7477979

CK14 719397 7478034

CK15 719506 7478023

HOMESTEAD HWY HH01 742071 7473652 ALICE'S BURROWS A03 720357 7481921 A07 720216 7481894

A09 720318 7481884

LAKES TOUR RD LTR01 712775 7482368 LTR12 712680 7482352

LTR13 712639 7482329

LTR14 712735 7482259

LATZY'S BREAK LB09 732326 7476010 LB12 732336 7476122

NYRIPPI SITES NY09 680439 7492327 NY15 680196 7492689

! 204! ! ! ! ! ! !

Fig. A1. Map of active Liopholis kintorei burrow systems located at the main study site. Groups identified in Chapter 3 are named after the burrow in which they were captured. WGS84 UTMs are given on the axes: eastings (x-axis) and northings (y-axis)

! !

! 205! ! ! ! ! ! !

a) 25" 2011)

20" total"captures" 15" recaptures" 10" Frequency)

5"

0"

b) 25" 2012)

20"

15"

10" frequency)

5"

0"

c) 25" 2013)

20"

15"

10" frequency)

5"

0"

Fig. A2. Histogram of lizard captures and the proportion of within-season recaptures for each successful trapping night during the (a) 2011, (b) 2012, and (c) 2013 field seasons

! 206!

01.11.13"

31.10.13"

30.10.13"

28.10.13"

27.10.13"

25.10.13"

24.10.13"

23.10.13"

21.10.13"

18.10.13"

17.10.13"

13.10.13"

11.10.13"

09.10.13"

07.10.13"

08.01.13" 07.01.13"

of study of 05.01.13"

04.01.13"

31.12.12"

30.12.12"

27.12.12"

23.12.12"

16.12.12"

14.12.12"

07.11.12"

06.11.12"

03.11.12"

31.10.12"

30.10.12"

28.10.12"

27.10.12"

20.10.12"

19.10.12"

16.10.12"

15.10.12"

13.10.12"

12.10.12"

11.10.12"

09.10.12"

08.10.12"

29.10.11"

27.10.11"

26.10.11"

25.10.11"

24.10.11"

23.10.11"

22.10.11"

21.10.11"

20.10.11"

19.10.11" 18.10.11"

the cumulative proportion of recaptures per successful trapping night over all three years

16.10.11"

15.10.11"

total"captures" recaptures" 12.10.11" 10.10.11" Histogram of Histogram

5" 0" .

25" 20" 15" 10" Frequency) A3

. Fig

207 a) 35"

30"

25"

20"

15" No.)captures) 10"

5"

0" L0101" L0103" L0501" L0503" L0507" L0508" L0601" L0701" L0801" L1101" L1305" L1703" L1801" L2401" L2405" L2406" L4301" L4303" L4401" L0204" L0502" L0504" L0902" L1402" L1702" L1902" L2404" L4502"

b) 35"

30"

25"

20"

15" No.)captures) 10"

5"

0" L0201" L0401" L0803" L0901" L1501" L1601" L1701" L2301" L3601" L5201" L0202" L0702" L1304" L1602" L2402" L4302"

c) 35"

30"

25"

20"

15" No.)captures) 10"

5"

0" L0203" L0505" L0506" L0509" L1001" L1301" L1303" L1306" L1401" L1403" L1603" L1901" L2403" L2407" L4501" L4503" L0102" L0510" L0802" L1302" L1704"

Fig. A4. Histograms of the number of captures per individual lizard throughout the study for (a) females, (b) males, and (c) juveniles. Individual IDs are given on the X-axis

208! Table A6. Burrow occupancy – list of all captures throughout the study

BURROW ID DATE (MORNING) INDIVIDUAL ID

L01 8.10.12 L0103 (C5D9) L01 8.10.12 L0701 (4A33) L01 9.10.12 L0803 (4F58) L01 15.10.12 L0803 (4F58) L01 16.10.12 L0803 (4F58) L01 16.10.12 L0103 (C5D9) L01 27.10.12 L0801 (3D53) L01 27.10.12 L0803 (4F58) L01 27.10.12 L0701 (4A33) L01 19.10.12 L0103 (C5D9) L01 19.10.12 L0701 (4A33) L01 20.10.12 L0103 (C5D9) L01 20.10.12 L0701 (4A33) L01 28.10.12 L0701 (4A33) L01 28.10.12 L0803 (4F58) L01 30.10.12 L0701 (4A33) L01 30.10.12 L0803 (4F58) L01 31.10.12 L0803 (4F58) L01 3.11.12 L0701 (4A33) L01 3.11.12 L0803 (4F58) L01 6.11.12 L0103 (C5D9) L01 6.11.12 L0701 (4A33) L01 7.11.12 L0103 (C5D9) L01 8.11.12 L0701 (4A33) L01 8.11.12 L0803 (4F58) L01 8.11.12 L0103 (C5D9) L01 14.12.12 L0803 (4F58) L01 16.12.12 L0701 (4A33) L01 23.12.12 L0803 (4F58) L01 23.12.12 L0701 (4A33) L01 27.12.12 L0701 (4A33) L01 30.12.12 L0701 (4A33) L01 4.1.13 L0803 (4F58) L01 4.1.13 L0103 (C5D9) L01 4.1.13 L0701 (4A33) L01 7.1.13 L0701 (4A33) L02 11.10.12 L0202 (4CDC) L02 13.10.12 L0202 (4CDC) L02 15.10.12 L0202 (4CDC) L02 20.10.12 L0202 (4CDC) L02 28.10.12 L0202 (4CDC) L02 30.10.12 L0202 (4CDC) L02 3.11.12 L0202 (4CDC) L02 6.11.12 L0202 (4CDC)

209! BURROW ID DATE (MORNING) INDIVIDUAL ID L02 7.11.12 L0202 (4CDC) L02 12.11.12 L0202 (4CDC) L02 14.12.12 L0202 (4CDC) L02 14.12.12 L0101 (294F) L02 27.12.12 L0202 (4CDC) L02 27.12.12 L0204 (E375) L02 4.1.13 L0202 (4CDC) L02 5.1.13 L0202 (4CDC) L02 6.1.13 L0202 (4CDC) L02 8.1.13 L2405 (E5EE) L05 11.10.12 L0505 (231C) L05 11.10.12 L0501 (8C62) L05 12.10.12 L0503 (3ED6) L05 12.10.12 L0501 (8C62) L05 13.10.12 L0501 (8C62) L05 13.10.12 L0503 (3ED6) L05 13.10.12 L0502 (CE23) L05 15.10.12 L0503 (3ED6) L05 15.10.12 L0505 (231C) L05 15.10.12 L0501 (8C62) L05 16.10.12 L0502 (CE23) L05 27.10.12 L0505 (231C) L05 28.10.12 L0505 (231C) L05 30.10.12 L0501 (8C62) L05 30.10.12 L0505 (231C) L05 30.10.12 L0503 (3ED6) L05 31.10.12 L0507 (CFD7) L05 3.11.12 L0503 (3ED6) L05 3.11.12 L0505 (231C) L05 3.11.12 L0502 (CE23) L05 3.11.12 L0501 (8C62) L05 6.11.12 L0505 (231C) L05 7.11.12 L0507 (CFD7) L05 7.11.12 L0505 (231C) L05 11.11.12 L0501 (8C62) L05 11.11.12 L0504 (DF33) L05 11.11.12 L0505 (231C) L05 11.11.12 L0507 (CFD7) L05 11.11.12 L0503 (3ED6) L05 12.11.12 L0501 (8C62) L05 12.11.12 L0504 (DF33) L05 12.11.12 L0507 (CFD7) L05 12.11.12 L0505 (231C) L05 13.11.12 L0503 (3ED6) L05 13.11.12 L0501 (8C62) L05 13.11.12 L0504 (DF33)

210! ! !

BURROW ID DATE (MORNING) INDIVIDUAL ID L05 13.11.12 L0507 (CFD7) L05 13.11.12 L0505 (231C) L05 16.12.12 L0507 (CFD7) L05 16.12.12 L0501 (8C62) L05 16.12.12 L0505 (231C) L05 16.12.12 L0508 (0CEE) L05 27.12.12 L0508 (0CEE) L05 27.12.12 L0505 (231C) L05 27.12.12 L0501 (8C62) L05 31.12.12 L0505 (231C) L05 31.12.12 L0508 (0CEE) L05 31.12.12 L0509 (ED6A) L05 5.1.13 L0505 (231C) L05 5.1.13 L0508 (0CEE) L05 7.1.13 L0901 (F1B1) L05 7.1.13 L0508 (0CEE) L05 7.1.13 L0509 (ED6A) L05 7.1.13 L0507 (CFD7) L05 7.1.13 L0505 (231C) L05 8.1.13 L0508 (0CEE) L05 8.1.13 L0509 (ED6A) L05 8.1.13 L0507 (CFD7) L05 8.1.13 L0505 (231C) L05 9.1.13 L0508 (0CEE) L05 9.1.13 L0509 (ED6A) L05 9.1.13 L0507 (CFD7) L05 9.1.13 L0505 (231C) L06 30.10.12 L0601 (4454) L06 31.10.12 L0601 (4454) L06 3.11.12 L0601 (4454) L06 6.11.12 L0601 (4454) L06 7.11.12 L0601 (4454) L06 16.12.12 L0601 (4454) L06 31.12.12 L0601 (4454) L08 8.10.12 L0101 (294F) L08 9.10.12 L0802 (7905) L08 15.10.12 L0802 (7905) L08 16.10.12 L0101 (294F) L08 16.10.12 L0802 (7905) L08 20.10.12 L0802 (7905) L08 28.10.12 L0103 (C5D9) L08 31.10.12 L0101 (294F) L08 3.11.12 L0101 (294F) L08 6.11.12 L0101 (294F) L08 7.11.12 L0101 (294F) L08 18.11.12 L0103 (C5D9)

! 211! ! ! ! BURROW ID DATE (MORNING) INDIVIDUAL ID L08 21.11.12 L0103 (C5D9) L08 21.11.12 L0701 (4A33) L08 21.11.12 L0803 (4F58) L08 16.12.12 L0103 (C5D9) L09 30.10.12 L0902 (550F) L09 7.11.12 L0902 (550F) L09 16.12.12 L0902 (550F) L09 31.12.12 L0902 (550F) L09 5.1.13 L0902 (550F) L09 8.1.13 L0902 (550F) L11 27.10.12 L1101 (AEEB) L11 30.10.12 L1101 (AEEB) L11 31.10.12 L1101 (AEEB) L12 13.10.12 L1701 (2B98) L12 15.10.12 L1701 (2B98) L12 19.10.12 L1701 (2B98) L12 20.10.12 L1701 (2B98) L12 3.11.12 L1701 (2B98) L13 9.10.12 L1601 (4ED6) L13 9.10.12 L1305 (3667) L13 9.10.12 L1304 (3A17) L13 12.10.12 L1305 (3667) L13 12.10.12 L1303 (68A7) L13 15.10.12 L1304 (3A17) L13 15.10.12 L1303 (68A7) L13 16.10.12 L1303 (68A7) L13 16.10.12 L1304 (3A17) L13 16.10.12 L1305 (3667) L13 17.10.12 L1303 (68A7) L13 17.10.12 L1304 (3A17) L13 17.10.12 L1305 (3667) L13 19.10.12 L1303 (68A7) L13 19.10.12 L1304 (3A17) L13 19.10.12 L1303 (68A7) L13 19.10.12 L1304 (3A17) L13 20.10.12 L1305 (3667) L13 20.10.12 L1304 (3A17) L13 20.10.12 L1303 (68A7) L13 27.10.12 L1304 (3A17) L13 30.10.12 L1304 (3A17) L13 30.10.12 L1305 (3667) L13 31.10.12 L1303 (68A7) L13 3.11.12 L1304 (3A17) L13 7.11.12 L1305 (3667) L13 15.11.12 L1304 (3A17) L13 15.11.12 L1305 (3667)

212! ! !

BURROW ID DATE (MORNING) INDIVIDUAL ID L13 16.11.12 L1304 (3A17) L13 16.11.12 L1305 (3667) L13 17.11.12 L1303 (68A7) L13 17.11.12 L1304 (3A17) L13 17.11.12 L1305 (3667) L13 14.12.12 L1303 (68A7) L13 23.12.12 L1303 (68A7) L13 23.12.12 L1801 (45F2) L13 28.12.12 L1303 (68A7) L13 28.12.12 L1801 (45F2) L13 30.12.12 L1801 (45F2) L13 30.12.12 L1304 (3A17) L13 31.12.12 L1303 (68A7) L13 31.12.12 L1304 (3A17) L13 4.1.13 L1801 (45F2) L13 4.1.13 L1303 (68A7) L13 4.1.13 L1304 (3A17) L13 7.1.13 L1303 (68A7) L14 8.10.12 L2402 (DE3B) L14 8.10.12 L1602 (4A45) L14 15.10.12 L1602 (4A45) L14 15.10.12 L2402 (DE3B) L14 16.10.12 L2402 (DE3B) L14 16.10.12 L1602 (4A45) L14 19.10.12 L2402 (DE3B) L14 19.10.12 L1602 (4A45) L14 27.10.12 L1602 (4A45) L14 28.10.12 L1602 (4A45) L14 30.10.12 L1602 (4A45) L14 31.10.12 L1602 (4A45) L14 14.12.12 L1602 (4A45) L15 3.11.12 L1402 (CCBF) L15 6.11.12 L1402 (CCBF) L15 30.12.12 L1501 (D809) L15 30.12.12 L1402 (CCBF) L15 4.1.13 L1501 (D809) L16 23.12.12 L1601 (4ED6) L16 30.12.12 L0202 (4CDC) L16 30.12.12 L1601 (4ED6) L16 4.1.13 L0204 (E375) L17 9.10.12 L1702 (E79D) L17 12.10.12 L1702 (E79D) L17 15.10.12 L1702 (E79D) L17 19.10.12 L1702 (E79D) L17 27.10.12 L1702 (E79D) L17 28.10.12 L1702 (E79D)

! 213! ! ! ! ! !

BURROW ID DATE (MORNING) INDIVIDUAL ID L17 31.10.12 L1702 (E79D) L17 31.10.12 L1703 (DED6) L17 3.11.12 L1702 (E79D) L17 3.11.12 L1703 (DED6) L17 4.11.12 L1702 (E79D) L17 4.11.12 L1703 (DED6) L17 5.11.12 L1702 (E79D) L17 5.11.12 L1703 (DED6) L17 6.11.12 L1702 (E79D) L17 6.11.12 L1703 (DED6) L17 7.11.12 L1702 (E79D) L17 7.11.12 L1703 (DED6) L17 6.12.12 L1703 (DED6) L17 14.12.12 L1703 (DED6) L17 23.12.12 L1703 (DED6) L17 30.12.12 L1703 (DED6) L17 4.1.13 L1703 (DED6) L17 7.1.13 L1703 (DED6) L18 16.10.12 L1304 (3A17) L18 17.10.12 L1304 (3A17) L18 7.11.12 L0201 (3910) L18 14.12.12 L1801 (45F2) L18 30.12.12 L1303 (68A7) L19 9.10.12 L0201 (3910) L19 13.10.12 L1304 (3A17) L19 20.10.12 L0201 (3910) L19 3.11.12 L1902 (F602) L19 7.11.12 L1304 (3A17) L19 14.12.12 L1304 (3A17) L19 23.12.12 L1902 (F602) L19 30.12.12 L1902 (F602) L19 4.1.13 L0201 (3910) L20 15.10.12 L0702 (DE06) L20 20.10.12 L0503 (3ED6) L20 20.10.12 L0501 (8C62) L20 20.10.12 L0702 (DE06) L20 27.10.12 L0702 (DE06) L20 28.10.12 L0503 (3ED6) L20 6.11.12 L0503 (3ED6) L20 7.11.12 L0503 (3ED6) L20 16.12.12 L0503 (3ED6) L20 31.12.12 L0507 (CFD7) L20 5.1.13 L0507 (CFD7) L23 13.10.12 L1601 (4ED6) L23 13.10.12 L2405 (E5EE) L23 15.10.12 L2405 (E5EE)

! 214! ! ! ! BURROW ID DATE (MORNING) INDIVIDUAL ID L23 15.10.12 L1601 (4ED6) L23 16.10.12 L2405 (E5EE) L23 16.10.12 L1601 (4ED6) L23 20.10.12 L2405 (E5EE) L23 20.10.12 L2301 (9E66) L23 28.10.12 L2405 (E5EE) L23 28.10.12 L2301 (9E66) L23 30.10.12 L2301 (9E66) L23 31.10.12 L2405 (E5EE) L23 31.10.12 L2301 (9E66) L23 1.11.12 L2301 (9E66) L23 2.11.12 L2301 (9E66) L23 3.11.12 L2405 (E5EE) L23 7.11.12 L2405 (E5EE) L23 18.11.12 L2301 (9E66) L23 20.11.12 L2301 (9E66) L23 20.11.12 L2405 (E5EE) L23 21.11.12 L2301 (9E66) L23 21.11.12 L2405 (E5EE) L23 16.12.12 L2405 (E5EE) L23 16.12.12 L2301 (9E66) L23 24.12.12 L2405 (E5EE) L23 24.12.12 L2301 (9E66) L23 25.12.12 L2301 (9E66) L23 25.12.12 L1601 (4ED6) L23 25.12.12 L2406 (D857) L23 27.12.12 L2405 (E5EE) L23 27.12.12 L2301 (9E66) L23 31.12.12 L1601 (4ED6) L23 5.1.13 L2301 (9E66) L23 5.1.13 L2405 (E5EE) L24 8.10.12 L2407 (31B2) L24 8.10.12 L2406 (D857) L24 8.10.12 L2401 (495E) L24 9.10.12 L2407 (31B2) L24 9.10.12 L2404 (3995) L24 9.10.12 L2406 (D857) L24 11.10.12 L2404 (3995) L24 12.10.12 L2406 (D857) L24 12.10.12 L2407 (31B2) L24 15.10.12 L2406 (D857) L24 16.10.12 L2406 (D857) L24 16.10.12 L2404 (3995) L24 19.10.12 L1601 (4ED6) L24 19.10.12 L2407 (31B2) L24 19.10.12 L2406 (D857)

215! BURROW ID DATE (MORNING) INDIVIDUAL ID L24 19.10.12 L2404 (3995) L24 20.10.12 L2404 (3995) L24 20.10.12 L2406 (D857) L24 20.10.12 L1601 (4ED6) L24 27.10.12 L2407 (31B2) L24 27.10.12 L2406 (D857) L24 28.10.12 L2407 (31B2) L24 30.10.12 L2404 (3995) L24 30.10.12 L2406 (D857) L24 31.10.12 L2407 (31B2) L24 31.10.12 L1601 (4ED6) L24 3.11.12 L2407 (31B2) L24 3.11.12 L2406 (D857) L24 3.11.12 L1601 (4ED6) L24 3.11.12 L2401 (495E) L24 7.11.12 L2407 (31B2) L24 8.11.12 L2401 (495E) L24 8.11.12 L2407 (31B2) L24 8.11.12 L1601 (4ED6) L24 8.11.12 L2406 (D857) L24 9.11.12 L2401 (495E) L24 9.11.12 L2407 (31B2) L24 14.12.12 L2407 (31B2) L24 14.12.12 L1601 (4ED6) L24 16.12.12 L2407 (31B2) L24 16.12.12 L2406 (D857) L24 17.12.12 L2401 (495E) L24 17.12.12 L2407 (31B2) L24 23.12.12 L2406 (D857) L24 27.12.12 L2407 (31B2) L24 27.12.12 L2406 (D857) L24 27.12.12 L1601 (4ED6) L24 30.12.12 L2407 (31B2) L24 30.12.12 L2401 (495E) L24 30.12.12 L2406 (D857) L24 4.1.13 L2407 (31B2) L24 4.1.13 L2406 (D857) L24 7.1.13 L2407 (31B2) L24 7.1.13 L1601 (4ED6) L32 30.10.12 L0502 (CE23) L32 30.10.12 L0504 (DF33) L32 31.10.12 L0502 (CE23) L32 6.11.12 L0502 (CE23) L32 7.11.12 L0502 (CE23) L32 16.12.12 L0502 (CE23) L32 27.12.12 L0502 (CE23)

216! BURROW ID DATE (MORNING) INDIVIDUAL ID L32 31.12.12 L0502 (CE23) L32 5.1.13 L0502 (CE23) L32 8.1.13 L0502 (CE23) L33 7.11.12 L1602 (4A45) L36 27.10.12 L1402 (CCBF) L36 30.10.12 L1402 (CCBF) L36 31.10.12 L1402 (CCBF) L36 3.11.12 L3601 (9C03) L36 6.11.12 L3601 (9C03) L36 7.11.12 L1402 (CCBF) L36 14.12.12 L2401 (495E) L36 23.12.12 L1402 (CCBF) L36 30.12.12 L3601 (9C03) L36 4.1.13 L1601 (4ED6) L36 4.1.13 L1402 (CCBF) L36 7.1.13 L1402 (CCBF) L43 31.10.12 L4301 (F66E) L43 14.12.12 L4302 (AD27) L43 16.12.12 L4301 (F66E) L43 23.12.12 L4301 (F66E) L43 30.12.12 L4301 (F66E) L43 4.1.13 L4301 (F66E) L44 30.10.12 L1701 (2B98) L44 31.10.12 L1701 (2B98) L44 14.12.12 L1701 (2B98) L44 23.12.12 L1701 (2B98) L44 4.1.13 L1701 (2B98) L44 7.1.13 L1701 (2B98) L45 27.12.12 L4501 (DDA9) L45 31.12.12 L4501 (DDA9) L45 31.12.12 L4502 (EC91) L45 5.1.13 L4502 (EC91) L45 8.1.13 L4503 (87C1) L45 8.1.13 L4501 (DDA9) L45 8.1.13 L4502 (EC91) L48 23.12.12 L0201 (3910) L48 30.12.12 L0201 (3910) L48 4.1.13 L1902 (F602) L48 7.1.13 L0201 (3910) L50 7.1.13 L1304 (3A17) L52 30.12.12 L5201 (18F8) L52 4.1.13 L5201 (18F8)

217!

- - - - 2 2 1 2 2 2 1 1 3 2 2 2 2 1 1 2 2 1 2 2 1 size size classes no. scat scat no.

- - - - 0 2 1 2 1 0 1 1 2 1 1 0 1 3 0 2 1 1 3 0 1 mounds no. basking basking no.

- - - - 2 1 1 1 1 1 1 2 2 1 2 1 1 1 1 1 1 1 1 1 1 no. latrines no.

- - - - 8 5 7 3 5 6 6 5 4 8 9 5 4 4 3 3 17 12 12 13 10 no. tunnels no.

–

- - - - 8 8 10 9.8 6.7 8.2 7.4 8.5 9.2 5.6 5.6 8.4 5.8 6.3 5.8 7.4 8.7 8.8 9.1 11.5 11.7 widest widest Span Span point (m) point

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 40 20 60 20

>1m cover cover % shrub shrub %

40 50 30 20 20 20 30 30 50 20 70 30 40 60 20 50 20 50 20 30 40 20 30 20 40 % ground ground % cover <1m cover

60 50 30 80 80 60 10 70 50 80 30 70 60 40 80 50 80 50 80 50 60 80 70 80 60 % bare bare % ground

4 4 1 2 4.4 2.4 5.5 3.3 5.7 3.4 5.5 2.9 6.2 6.3 3.3 1.6 1.3 6.6 3.8 5.6 3.7 0.5 3.8 5.3 4.9 termites distance to to distance

0 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 burrow (1) (1) burrow (0) control Characteristics of burrow systems and control quadrats (Chapter 3) (Chapter quadrats control and systems burrow of Characteristics . 7

ID C1 C2 C3 C4 L24 L32 L36 L43 L44 L17 L18 L19 L20 L23 L11 L12 L13 L14 L15 L01 L02 L05 L06 L08 L09 quadrat quadrat Table A Table

218

------size size classes no. scat scat no.

ng ------mounds no. baski no.

------no. latrines no.

------no. tunnels no.

–

------widest widest Span Span point (m) point

0 0 0 0 60 40 10 40 20 60 50 10 25 10 40 60 >1m cover cover % shrub shrub %

10 10 20 10 10 20 10 10 30 30 10 25 10 20 40 10 % ground ground % cover <1m cover

20 50 70 50 70 20 90 40 70 70 80 50 80 80 20 30 % bare bare % ground

8 2 6.6 1.8 3.6 4.4 6.6 6.9 2.6 7.5 1.6 2.6 7.7 3.2 10.1 13.1 termites distance to to distance

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 burrow (1) (1) burrow (0) control

ID C7 C8 C9 C5 C6 C19 C20 C21 C12 C14 C15 C16 C17 C18 C10 C11 quadrat quadrat

219

122 114 114 122 122 118 118 118 118 110 122 122 114 122 122 122 110 110 110 122 118 114 118 118 114 122

FQR 114 110 110 118 118 110 114 110 110 110 110 110 110 122 110 110 110 110 110 110 110 110 110 114 110 118

238 238 238 238 238 238 238 238 238 222 218 238 238 238 238 238 234 222 222 222 242 242 242 238 242 238

CKD 218 218 238 218 238 238 218 238 218 218 218 238 238 218 238 238 234 218 218 218 230 234 238 230 218 238

144 144 144 144 144 144 144 144 144 160 152 144 152 144 144 144 160 152 144 144 172 144 172 172 144 144

BX6 144 144 144 144 144 144 144 144 144 152 120 144 144 120 132 144 152 120 144 120 160 120 156 160 140 144

0 295 295 287 295 295 295 287 283 287 295 275 275 275 271 275 295 275 275 271 275 275 271 271 275 287

EST9 0 283 283 275 287 283 287 283 275 275 263 267 275 263 263 263 275 271 271 267 271 263 263 255 267 275

320 320 320 320 320 320 320 320 320 328 328 320 332 320 320 320 340 328 328 320 328 328 324 328 324 320

EST12 320 320 308 308 320 320 308 320 320 320 320 320 320 320 320 320 340 320 320 308 320 316 320 324 320 320

265 257 265 265 265 265 261 261 261 265 265 265 265 261 265 265 269 265 265 269 265 269 265 269 261 261

ECU3 253 253 249 253 253 253 261 261 249 261 257 245 265 249 249 249 265 249 249 265 249 249 249 253 249 249

0 0

152 152 152 152 152 144 152 152 152 144 144 144 144 144 152 144 144 152 144 144 144 144 152 152

J3F 0 0 144 144 152 144 144 144 152 152 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144

0 240 240 272 240 220 236 272 272 272 236 256 244 244 244 236 236 236 236 236 236 236 236 240 272 272

EST2 0

236 236 236 236 212 224 272 272 220 220 256 236 220 240 212 220 212 212 212 220 212 236 236 260 220

291 283 387 263 263 343 387 391 387 311 395 311 327 303 311 311 295 295 295 311 295 311 311 295 303 303 samples

EST1 263 227 227 227 227 319 291 291 303 311 275 307 307 279 227 227 263 263 263 263 227 295 263 263 303 291

161 161 161 161 161 161 161 161 161 161 161 175 161 161 161 161 161 161 161 161 161 161 161 161 161 161

ECU2 Liopholis kintorei Liopholis 161 157 161 161 161 161 161 161 161 161 161 157 161 157 161 161 161 161 157 157 161 161 161 161 157 157

SITE Genotypes for all all for Genotypes Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven . 8

Table A Table L0802 L0803 L0901 L0902 L0510 L0601 L0701 L0702 L0801 L0505 L0506 L0507 L0508 L0509 L0401 L0501 L0502 L0503 L0504 L0103 L0201 L0202 L0203 L0204 L0101 L0102 Sample ID Sample

220

110 114 114 114 110 110 114 118 110 110 110 110 114 114 110 114 122 114 118 114 118 118 122 114 114 110 122 122

FQR 110 110 114 110 110 110 110 114 110 110 110 110 110 110 110 110 122 114 114 110 114 114 118 114 110 110 110 110

234 242 234 242 234 234 234 234 234 222 238 234 242 242 234 234 238 246 234 234 234 234 222 234 242 238 238 234

CKD 222 222 222 230 234 234 222 210 222 218 234 234 234 218 222 234 222 238 214 222 218 218 218 234 230 222 222 222

140 160 160 172 144 160 144 140 160 144 160 140 144 144 144 152 144 152 160 140 144 144 152 156 172 144 152 152

BX6 140 156 156 160 140 144 120 120 144 144 152 140 120 140 140 144 144 144 152 120 140 140 144 140 160 144 144 140

0 267 275 275 271 275 271 275 275 267 283 287 275 287 271 271 275 267 275 275 271 275 275 283 275 271 267 271

EST9 0 267 263 275 263 271 267 267 275 263 275 283 267 263 267 263 271 263 267 271 271 271 271 271 271 267 267 267

320 320 324 324 336 336 328 336 332 328 328 336 332 320 336 336 320 332 336 324 324 324 324 332 328 320 320 320

EST12 316 320 320 320 328 328 328 328 324 324 324 336 316 320 320 336 320 320 336 308 324 324 320 324 320 320 320 316

261 277 253 265 257 265 265 265 265 265 253 265 265 249 249 265 253 261 265 265 265 265 265 265 261 253 265 261

ECU3 257 253 249 245 245 249 257 261 245 245 245 265 245 245 245 245 249 245 261 245 265 265 253 265 261 253 253 249

0

144 144 152 172 152 152 152 152 152 152 144 144 144 172 152 152 144 144 144 144 152 152 152 152 152 152 144

J3F 0 144 144 144 152 152 144 144 144 144 152 144 144 144 144 152 144 144 144 144 144 144 144 144 144 152 144 144

0 0 0 244 236 232 240 240 272 232 228 228 244 232 244 240 240 240 240 236 244 228 272 272 272 232 260 240

EST2 0 0 0 240 220 220 220 220 236 220 220 220 220 228 240 232 232 236 236 232 236 220 220 220 260 220 232 240

291 387 387 291 291 303 307 311 303 299 311 295 323 263 263 387 303 351 351 323 303 303 303 303 303 387 307 351

EST1 259 287 263 259 259 263 299 295 295 227 299 227 311 259 259 263 263 311 227 311 299 303 303 303 251 307 275 275

161 171 161 171 161 175 175 171 175 161 171 171 161 161 161 161 161 161 161 161 161 161 171 171 161 161 161 161

ECU2 161 157 157 161 161 161 157 171 157 161 161 171 159 157 161 161 157 157 161 161 161 161 161 161 157 161 161 161

SITE Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven

L2402 L2403 L2404 L2405 L1801 L1901 L1902 L2301 L2401 L1603 L1701 L1702 L1703 L1704 L1402 L1403 L1501 L1601 L1602 L1303 L1304 L1305 L1306 L1401 L1001 L1101 L1301 L1302 Sample ID Sample

221

114 122 114 118 114 114 114 122 118 114 122 122 114 122 114 122 114 114 114 122 110 122 122 118 118 122 122 122

FQR 110 110 114 110 110 110 110 114 114 114 118 114 110 114 110 122 114 110 110 110 110 114 114 114 110 118 110 110

0

250 234 254 238 222 246 238 254 254 254 238 218 242 238 242 254 234 238 238 238 234 218 218 210 234 210 234

CKD 0 218 218 218 234 218 218 206 226 234 234 238 214 234 238 214 206 222 238 238 222 218 218 210 210 210 210 218

172 156 152 152 160 152 152 156 156 172 144 144 144 152 144 152 140 152 152 144 160 160 140 144 172 172 160 144

BX6 140 152 152 144 140 144 144 140 132 152 144 144 144 152 144 140 140 144 120 140 152 160 140 140 160 152 152 140

0 0 275 283 267 295 295 279 299 291 291 295 271 271 279 275 275 275 275 275 299 271 307 283 275 275 275 275

EST9 0 0 267 255 267 255 291 267 279 283 255 275 267 267 275 275 263 267 271 263 263 263 271 275 271 271 271 271

0 312 340 336 332 340 324 328 340 340 332 344 324 324 324 336 336 336 336 324 340 316 316 320 336 324 336 316

EST12 0 308 328 316 328 324 320 324 340 320 324 336 316 324 316 324 336 332 320 320 328 316 316 300 324 324 324 316

261 253 253 253 261 269 265 261 269 273 253 273 269 269 265 253 261 281 265 281 257 265 273 265 261 265 261 265

ECU3 261 253 253 249 253 261 257 253 253 253 245 265 265 249 249 245 257 265 261 277 245 265 253 261 261 261 257 257

152 172 168 144 152 148 148 152 172 152 152 144 152 152 144 152 144 144 144 144 144 172 144 144 144 144 144 152

J3F 144 168 144 144 144 144 144 144 144 152 152 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144

260 272 224 272 240 260 256 256 264 220 220 240 260 260 240 244 236 248 280 232 256 268 268 280 256 264 256 244

EST2 240 272 224 272 240 220 248 236 244 204 204 216 220 240 220 204 236 228 240 228 232 236 220 248 248 240 232 220

0 275 303 271 303 275 327 323 259 287 387 323 299 311 311 387 255 307 363 307 279 295 291 303 275 307 335 291

EST1 0 251 271 259 271 271 251 303 227 275 259 311 287 287 295 259 227 251 239 275 263 279 287 275 263 243 247 227

0 161 161 161 161 161 161 159 161 175 161 175 161 171 161 171 157 175 161 161 161 161 171 161 161 161 157 157

ECU2 0 161 161 161 157 161 161 157 161 161 161 161 161 157 157 161 157 157 157 157 159 161 161 161 161 157 157 157

SITE Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven

RK - L4503 L5201 L4302 L4303 L4401 L4501 L4502 L2406 L2407 L3601 L4301 C4501 C5101 C7801 C3801 C3802 C1 C20302 C20303 C20304 C20701 C10702 C11601 C20101 C20201 C20301 C10601 C10701 Sample ID Sample

222

122 118 122 114 114 118 118 114 114 110 114 118 118 118 110 122 110 114 122 110 122 114 122 118 122 114 114 122

FQR 118 114 114 110 110 114 114 110 110 110 110 110 110 114 110 110 110 114 118 110 110 114 118 114 118 110 114 118

234 242 230 246 234 234 234 242 246 218 246 226 226 222 234 238 234 254 238 222 218 238 238 230 250 222 218 218

CKD 222 218 214 218 234 218 234 234 234 218 226 218 222 214 210 222 230 254 218 218 218 214 230 214 238 214 214 218

152 144 172 160 160 160 160 160 160 172 172 172 140 160 152 160 160 172 144 160 152 172 152 172 152 144 172 144

BX6 140 132 152 152 160 144 144 156 160 152 144 120 120 160 152 120 156 140 144 152 120 144 144 152 128 120 144 144

0 267 271 275 283 271 271 271 275 271 271 271 271 275 271 271 275 275 279 279 283 279 279 275 275 279 267 271

EST9 0 267 267 271 271 271 267 271 271 271 267 271 271 271 271 271 275 275 271 267 279 271 271 271 271 275 263 271

0 0 336 356 336 344 344 320 332 344 316 308 336 316 332 316 324 340 336 332 340 336 340 324 356 320 328 340

EST12 0 0 324 324 324 320 320 308 332 320 304 304 304 304 308 300 320 308 332 320 324 332 324 324 340 320 324 332

281 277 261 277 261 261 265 281 269 277 273 281 269 281 261 273 261 281 277 261 269 265 261 257 269 265 265 261

ECU3 277 253 257 277 257 257 261 281 261 273 265 269 265 261 253 261 261 233 265 257 265 261 257 253 265 261 257 261

144 144 144 144 144 148 144 152 152 144 152 152 144 152 152 152 152 144 152 144 144 144 144 152 144 144 144 148

J3F 144 144 144 144 144 144 144 144 144 144 144 144 144 152 152 152 144 144 144 144 144 144 144 144 144 144 144 148

0 268 236 240 224 268 268 244 260 260 256 256 260 248 252 248 252 256 228 256 248 252 264 252 248 260 240 260

EST2 0 236 232 220 220 256 252 236 236 220 220 228 232 248 248 248 248 248 220 256 240 248 248 244 240 232 232 260

0 0 283 283 283 315 295 283 303 303 299 303 303 323 311 303 307 307 271 303 291 291 291 275 295 295 303 299

EST1 0 0 279 279 239 291 283 275 299 271 251 251 251 287 307 303 303 251 251 287 251 267 267 271 271 295 279 251

161 161 175 161 175 175 175 171 171 175 169 161 161 161 175 161 161 157 175 161 175 175 161 161 161 161 161 161

ECU2 157 161 157 157 161 171 171 153 157 161 157 161 157 161 161 161 161 157 161 157 167 161 161 161 161 161 161 159

SITE Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven

A0901 A0301 A0302 A0303 A0701 C22901 C22401 C22402 C22501 C22701 C22801 C21101 C21102 C21103 C22201 C22202 C20702 C20801 C20802 C20803 NY0901 NY0902 HH0101 LTR1401 LTR0101 LTR0102 LTR1201 LTR1301 Sample ID Sample

223

0

118 114 118 114 114 114 118 114 114 114 122 114 114 118 122 118 114 118 118 114 118 114 118 118 118 118 118

FQR 0 114 114 110 114 110 114 114 114 110 114 114 110 110 114 114 114 110 114 114 110 114 114 110 114 114 110 114

238 218 242 238 250 234 242 242 234 242 210 222 250 230 226 230 242 230 234 238 238 238 238 238 238 222 246 234

CKD 222 218 222 238 246 230 230 218 230 210 206 222 234 202 222 218 230 226 210 210 210 234 222 238 214 210 214 222

0

160 152 172 144 172 172 140 152 144 152 140 172 172 172 152 172 136 144 172 152 140 144 156 156 156 144 144

BX6 0 140 144 140 132 152 144 132 144 132 144 132 132 152 172 144 140 120 140 144 144 140 120 156 144 152 144 120

0 271 299 271 291 271 271 275 279 279 295 287 271 295 279 275 287 283 279 299 283 287 295 271 279 279 279 295

EST9 0 263 275 271 291 267 263 271 267 255 295 271 267 271 279 267 271 283 271 271 271 283 263 267 271 275 271 263

0 340 336 320 320 328 340 324 336 332 336 336 324 332 324 328 332 328 336 328 344 324 320 332 332 320 332 320

EST12 0 336 328 320 316 324 332 316 332 316 324 336 312 320 320 316 316 316 320 316 320 324 316 320 320 320 324 316

277 265 257 265 277 257 273 273 265 265 261 273 277 265 261 265 273 273 257 273 265 261 257 261 277 273 269 285

ECU3 273 265 249 253 257 257 233 233 253 257 253 245 273 261 249 253 253 261 249 273 265 253 249 257 249 265 249 261

0

144 144 144 148 144 144 148 144 144 148 144 152 144 144 152 168 144 144 144 172 144 152 144 144 172 144 144

J3F 0 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 144 152 144 144

0 292 260 272 228 280 268 268 256 248 244 244 228 268 264 252 284 284 276 276 268 248 264 280 260 244 260 264

EST2 0 248 236 264 220 220 264 232 252 220 220 220 220 264 232 244 284 224 244 224 236 244 224 252 248 228 232 252

0 0 0 0 311 295 295 391 339 307 295 291 279 319 283 251 303 275 299 247 279 263 299 383 307 287 271 295

EST1 0 0 0 0 247 283 251 291 295 259 255 263 267 267 271 251 271 259 247 227 263 251 295 239 263 275 251 283

175 161 161 161 161 175 175 175 161 161 175 175 171 171 161 175 161 161 161 159 161 171 161 167 161 159 161 161

ECU2 169 157 157 161 157 161 161 157 159 157 161 161 161 159 161 171 157 157 157 157 157 161 159 161 157 157 161 157

SITE Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven

8699 9890 3452 3B89 679B 6B90 81AF 3AF5 651D A93E 8A6C A9EC B2CC B1001 B2801 B3001 LB0901 LB0902 LB1201 CK1401 CK1402 CK1501 CK0201 CK0801 CK0802 CK0803 CK1001 NY1501 Sample ID Sample

!224

122 114 110 110 122 110 110 110 110 122 110 122 110 110 110 110 110 114 114 114 122 114 114 114 118 114 110 122

FQR 110 110 110 110 110 110 110 110 110 110 110 114 110 110 110 110 110 114 110 110 110 110 110 114 114 106 110 110

0 0

222 242 222 250 230 246 234 246 238 246 230 238 242 242 246 246 238 234 246 246 246 246 250 230 234 238

CKD 0 0 206 234 218 230 210 234 226 246 206 222 218 230 222 226 210 234 230 230 238 234 234 210 230 218 214 238

160 140 148 156 160 144 144 156 160 160 168 148 156 148 148 148 148 152 144 156 152 152 172 152 144 156 148 156

BX6 140 132 144 148 156 110 140 140 156 160 140 140 148 140 140 140 148 144 140 152 140 152 144 140 128 136 132 136

0 0 0 0 279 287 279 279 291 279 287 291 283 291 267 271 291 279 271 271 283 283 267 267 271 271 271 287

EST9 0 0 0 0 275 271 267 259 275 271 275 279 275 283 259 263 279 259 267 271 271 275 255 267 267 263 267 287

0 0 0 332 332 320 344 324 308 328 332 324 352 320 328 320 340 324 328 320 324 332 344 336 320 324 352 324

EST12 0 0 0 320 324 320 320 312 304 324 328 320 312 316 308 316 324 312 324 308 316 316 324 332 316 304 312 312

0 261 277 265 273 277 277 281 265 261 277 269 273 269 257 293 269 261 281 269 257 269 221 257 261 253 257 285

ECU3 0 225 253 261 261 273 269 265 265 261 249 265 249 261 221 277 253 249 257 249 253 261 221 225 249 249 249 269

160 148 168 160 160 160 144 144 152 144 172 152 144 144 148 144 160 144 144 148 160 180 148 144 144 160 184 144

J3F 160 144 144 144 144 148 144 144 152 144 144 144 144 140 144 144 144 144 144 144 148 144 144 144 144 144 148 144

232 264 264 248 248 244 260 268 264 264 236 232 280 232 248 244 248 240 228 264 252 260 236 252 264 248 260 252

EST2 232 224 264 220 220 244 232 248 248 232 236 228 264 220 236 240 236 236 228 204 232 240 224 232 236 240 228 224

0 279 307 259 303 287 259 307 279 251 279 295 275 275 239 291 263 263 267 319 307 311 299 271 303 275 299 299

EST1 0 259 299 247 267 275 251 271 275 251 251 263 251 263 203 291 259 255 243 275 267 279 287 259 263 263 255 251

0 195 159 151 159 157 185 157 163 161 159 163 163 175 175 159 161 185 163 185 161 159 157 161 161 161 161 171

ECU2 0 157 157 151 159 151 171 151 157 157 151 163 151 161 161 159 161 151 157 157 159 159 157 157 159 161 157 161

SITE Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Newhaven Docker River Docker Sangster's Bore Sangster's Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's

SB16 SB17 SB18 SB19 SB11 SB12 SB13 SB14 SB15 SB06 SB07 SB08 SB09 SB10 SB01 SB02 SB03 SB04 SB05 BB33 DR01 B4401 B5001 B5601 B7901 B3201 B4101 B4102 Sample ID Sample

225

114 114 114 114 122 122 122 122 122 118 114 114 114 114 114 118 114 118 114 114 114 114 118 118 114 114 114 114

FQR 114 114 110 114 114 114 110 114 114 114 114 114 114 114 114 114 114 110 110 110 110 114 110 110 110 110 110 114

242 234 238 238 214 234 246 262 230 230 262 230 230 222 234 230 234 242 246 222 222 234 238 226 222 234 210 234

CKD 230 210 234 214 210 234 218 258 226 222 246 206 230 218 222 218 210 230 222 210 210 198 218 222 222 226 210 210

156 180 180 160 160 160 160 152 156 152 176 152 160 136 160 160 160 144 140 148 148 148 172 152 152 144 156 156

BX6 152 156 156 128 152 152 132 124 124 152 152 152 132 132 160 128 152 110 140 140 144 116 136 144 144 144 152 152

0 0 0 0 283 267 267 283 287 279 287 291 295 279 283 295 287 275 291 279 299 291 291 279 275 287 267 311

EST9 0 0 0 0 275 267 259 283 283 275 267 279 267 267 271 283 275 275 275 279 287 271 283 279 271 275 267 283

0 324 316 324 340 352 336 340 320 328 328 320 316 344 340 332 332 328 320 328 320 344 344 324 320 332 324 324

EST12 0 316 312 312 328 280 328 316 296 308 296 316 308 316 328 300 328 280 316 304 312 316 324 320 316 324 312 312

265 273 253 257 257 261 269 285 225 265 253 275 273 277 261 261 261 261 265 273 253 257 265 273 265 257 257 269

ECU3 249 257 245 253 249 261 269 269 225 253 253 273 257 265 257 257 261 257 257 261 245 245 253 257 257 253 257 257

148 148 144 148 148 148 172 172 172 144 144 144 148 144 144 144 148 144 144 152 148 148 144 148 148 144 148 144

J3F 144 144 144 148 144 144 168 144 168 144 144 144 144 144 144 144 144 144 144 144 144 144 144 148 144 144 144 144

284 244 256 252 252 232 256 256 256 248 248 252 224 252 280 280 236 248 256 248 268 240 288 244 284 256 264 272

EST2 244 240 256 236 228 228 240 224 240 236 248 236 224 236 236 248 236 236 236 248 236 240 236 240 256 240 228 268

0 0 295 323 275 319 271 275 211 211 211 275 299 291 299 299 299 295 295 291 231 267 299 271 295 299 315 263

EST1 0 0 271 291 263 271 231 267 211 211 211 251 267 239 291 295 295 291 259 283 231 259 259 255 259 271 251 247

193 193 189 191 193 189 189 165 171 189 165 189 187 175 171 183 181 171 169 161 191 169 199 159 163 153 171 189

ECU2 169 191 165 165 165 165 157 157 157 155 157 171 171 171 171 157 157 169 157 157 189 165 163 151 159 151 155 169

SITE Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Watarru Watarru Watarru Watarru Watarru Watarru Watarru Watarru Warburton Warburton Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's

U11 U12 U13 U14 U06 U07 U08 U09 U10 U01 U02 U03 U04 U05 SB20 SB21 SB22 SB23 WT04 WT05 WT06 WT07 WT08 WT01 WT02 WT03 WB01 WB02 Sample ID Sample

226

114 114 122 114 114 114 114 114 114 114 122 118 114 118 114 114

FQR 114 114 110 114 114 114 110 110 114 114 110 110 110 110 110 114

222 222 246 222 238 238 234 242 238 250 238 234 230 234 242 242

CKD 222 222 246 210 222 222 234 218 210 210 238 230 226 230 218 238

136 136 152 160 192 160 152 152 160 128 160 152 152 176 160 152

BX6 132 132 124 160 160 152 152 132 128 128 160 132 124 160 160 152

275 287 279 275 275 283 287 283 283 271 287 295 295 275 295 307

EST9 275 275 275 271 271 275 283 267 271 271 279 283 287 275 279 275

352 340 340 332 332 316 316 340 340 316 296 340 340 352 332 320

EST12 304 304 340 280 280 280 296 316 328 316 280 316 308 304 300 312

265 265 257 257 261 265 269 253 261 273 257 257 269 265 257 265

ECU3 253 265 253 257 249 257 265 245 253 269 253 253 257 257 253 257

148 144 152 152 148 148 144 144 148 144 144 144 144 148 148 148

J3F 144 144 144 148 144 144 144 144 144 144 144 144 144 144 144 144

248 288 256 284 248 284 288 288 252 236 256 256 256 244 244 244

EST2 220 236 248 248 240 244 284 236 248 236 256 256 240 228 244 244

271 299 291 283 275 287 295 279 271 271 299 271 271 319 303 303

EST1 251 275 251 251 259 267 267 271 271 255 275 263 263 267 259 259

189 189 203 165 161 203 189 189 169 169 169 165 189 191 157 169

ECU2 157 165 159 153 159 159 159 189 157 157 153 157 165 189 157 165

SITE Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru Uluru

U29 U30 U24 U25 U26 U27 U28 U19 U20 U21 U22 U23 U15 U16 U17 U18 Sample ID Sample

227

Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Bore Sangster's Uluru Uluru Warburton Watarru Locality Warburton Watarru, River, Docker Newhaven Newhaven Bore Sangster's Newhaven, Newhaven Newhaven Newhaven Bore Sangster's

...... T T C

...... T C

...... T C C C

...... T T C

. G G G G G G G G A G G G G G G G

...... G A G Liopholis kintorei Liopholis

...... T C

...... A G

...... T C C C C

...... G A A

...... A G

...... T C

...... A G G G G G G G G G

...... G A A A A

...... T C

...... T C

...... G A A A

...... G A

...... C T T T T T

...... T C

...... C A A A

...... G A

. G G G G G G G G G G G G G Variable sites Variable A G G

1 3 1 3 2 2 2 1 7 8 1 1 7 9 1 12 11 Frequency

Variable sites for 17 haplotypes sampled from six localities across the distribution of of distribution the across localities six from sampled haplotypes 17 for sites Variable

. 9 no. Hap_7 Hap_8 Hap_9 Hap_2 Hap_3 Hap_4 Hap_5 Hap_6 Hap_1 Hap_16 Hap_17 Hap_11 Hap_12 Hap_13 Hap_14 Hap_15 Hap_10 Haplotype Haplotype Table A Table

228 RESEARCH ARTICLE Genetic Divergence among Regions Containing the Vulnerable Great Desert Skink (Liopholis kintorei) in the Australian Arid Zone

Siobhan Dennison1,2*, Steve McAlpin1,3, David G. Chapple4, Adam J. Stow1

1 Department of Biological Sciences, Macquarie University, North Ryde, NSW, Australia, 2 Australian Wildlife Conservancy, PO Box 8070, Subiaco East, WA, Australia, 3 School of Environmental and Rural Science, University of New England, Armidale, NSW, Australia, 4 School of Biological Sciences, Monash University, Clayton, VIC, Australia

* [email protected]

Abstract

Knowledge of genetic structure and patterns of connectivity is valuable for implementation OPEN ACCESS of effective conservation management. The arid zone of Australia contains a rich biodiversi- Citation: Dennison S, McAlpin S, Chapple DG, Stow ty, however this has come under threat due to activities such as altered fire regimes, grazing AJ (2015) Genetic Divergence among Regions and the introduction of feral herbivores and predators. Suitable habitats for many species Containing the Vulnerable Great Desert Skink (Liopholis kintorei) in the Australian Arid Zone. PLoS can be separated by vast distances, and despite an apparent lack of current geographical ONE 10(6): e0128874. doi:10.1371/journal. barriers to dispersal, habitat specialisation, which is exhibited by many desert species, may pone.0128874 limit connectivity throughout this expansive region. We characterised the genetic structure Academic Editor: Bi-Song Yue, Sichuan University, and differentiation of the great desert skink (Liopholis kintorei), which has a patchy, but CHINA widespread distribution in the western region of the Australian arid zone. As a species of Received: February 10, 2015 cultural importance to local Aboriginal groups and nationally listed as Vulnerable, it is a

Accepted: May 2, 2015 conservation priority for numerous land managers in central Australia. Analysis of mitochon- drial ND4 sequence data and ten nuclear microsatellite loci across six sampling localities Published: June 10, 2015 through the distribution of L. kintorei revealed considerable differentiation among sites, with Copyright: © 2015 Dennison et al. This is an open mitochondrial F and microsatellite F ranging from 0.047-0.938 and 0.257-0.440, re- access article distributed under the terms of the ST 0ST Creative Commons Attribution License, which permits spectively. The extent of differentiation suggests three main regions that should be man- unrestricted use, distribution, and reproduction in any aged separately, in particular the southeastern locality of Uluru. Current genetic delineation medium, provided the original author and source are of these regions should be maintained if future intervention such as translocation or captive credited. breeding is to be undertaken. Data Availability Statement: Mitochondrial haplotype sequences from this study have been submitted to Genbank (accession numbers KM035773-KM035789). All other relevant data are within the paper and its Supporting Information files.

Funding: Most of the data collection and analysis of Introduction genetic material used in this study was funded by The Australian arid zone occupies 70% of the continent’s landmass and supports an extraordi- Macquarie University. Additional funding was nary biodiversity, including among the world’s richest assemblages of lizards [1], [2]. Despite a provided by the Joyce W. Vickery Fund (Linnean Society of NSW, to SD) and the Rice Memorial Fund longstanding recognition of the conservation value of this region, relatively few studies have (Macquarie University, to SD). In kind support for described patterns of genetic structuring across whole species distributions [3]. Characterisa- sample collection was provided by the Australian tion of genetic structure across a landscape is valuable to inform conservation because

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 1 / 14 Regional Genetic Structure of a Threatened Lizard

Wildlife Conservancy and Parks Australia. DGC genetically discrete regions may be under different pressures and require separate manage- received funding from the Monash University Faculty ment approaches. Connectivity among these regions may also not correspond to natural of Science Linkage Project Application Support boundaries as expected based on observed environmental or geographic features [4–6]. More- (LPAS) Scheme. The funders had no role in study design, data collection and analysis, decision to over, human land use is rapidly changing arid Australia and is posing a number of threats to publish, or preparation of the manuscript. the biodiversity of the region. Habitat destruction through land clearing, accelerated soil ero- sion, unsustainable cattle grazing and altered fire regimes continues to threaten inland Austra- Competing Interests: The authors have declared that no competing interests exist. lian biota, as do increased weed and feral animal populations [7–9]. Knowledge of levels of genetic connectivity can be used to evaluate the impact of localised activities and to prioritise conservation strategies. Although the arid regions of Australia currently lack substantial topographic barriers or ex- pansive waterways, genetic variation may be structured by environmental features. Many rep- tile species are habitat specialists, restricted to specific habitat types scattered throughout the desert region [10,11]. Vast expanses of contiguous habitat types such as dunefields or gibber plain may separate patches of suitable habitat for many species. This may result in habitat patches being isolated from each other by hundreds of kilometres, a distance in excess of the likely dispersal capacity of many desert species [11,12]. Thus, habitat heterogeneity, and the as- sociated habitat specialisation of arid zone lizards, might influence connectivity and restrict gene flow between localities. For example, Chapple et al. [12] found considerable phylogeo- graphic structure within Egernia inornata (now Liopholis inornata [13]), an arid-Australian lizard, with a number of clades occurring in particular habitat types. It is therefore our a priori expectation that localised impacts may threaten genetically distinct components of biodiversity. There are also additional benefits derived from knowledge of connectivity; for example, it is established that parts of a species distribution that experience prolonged isolation may become sufficiently genetically differentiated that they are worthy of separate management [14–16]. Furthermore, isolation coupled with reduced effective sizes can lower genetic diversity through drift and impinge on the ability to adapt to environmental change. In such cases, translocations among genetically discrete localities may not be a viable conservation strategy owing to the risk of outbreeding depression [16] (but see [17]). Identifying parts of the distribution requiring separate management enables conservation effort to be prioritized and can guide decisions to translocate, restore, or establish breeding programs [16,18]. The great desert skink, Liopholis kintorei, is a species endemic to the arid-zone of Australia, currently listed as ‘Vulnerable’ [19]. It is a large scincid lizard that inhabits sand plains, palaeo- drainage lines and undulating gravelly downs [20]. Although its range stretches over a vast area of approximately 1.3 million km2, it is known to be patchily distributed, with its presence recorded at fewer than 100 localities [21]. Great desert skinks exhibit limited dispersal (com- monly 0–4 km, up to 9km; [21]), excavating extensive burrow systems in which close kin live and which may be continuously occupied for up to 7 years [22]. It is a culturally important spe- cies to traditional Aboriginal groups [23], and this combined with its threatened status makes its conservation a high priority for land managers in central Australia. Altered fire regimes and the introduction of the red fox and domestic cat are key factors that have led to the species’ de- cline, as well as habitat decay from feral herbivores [8,19,20,24]. Areas containing great desert skinks are known to occur in a number of geographically dis- tant regions within declared conservation areas: Uluru-Kata-Tjuta National Park (NT), New- haven Wildlife Sanctuary (NT), Karlamilyi National Park (WA), Ngaanyatjarra Indigenous Protected Area (IPA; WA), and in the Watarru IPA within the Anangu Pitjantjatjara Yanku- nytjatjara Lands (APY; SA); and in these a number of monitoring and management actions have previously been undertaken. The extent of isolation or genetic differentiation between these regions is unknown, but likely to be high given the apparent disjunct distribution and

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 2 / 14 Regional Genetic Structure of a Threatened Lizard

low dispersal of the species. Here we use mtDNA sequence data (ND4) and ten microsatellite loci to characterise genetic structure and divergence across the range of L. kintorei. We aim to identify any regions where the genetic distinctiveness of L. kintorei heightens the conservation value and influences the management options.

Materials and Methods Ethics Statement All animals were handled in accordance with a protocol considered and approved by Mac- quarie and Charles Darwin University Animal Ethics committee recommendations (ARA 2008/025 and ARA 2011/037). Sample collection was licensed by the Northern Territory Parks and Wildlife Commission and the South Australian Department of Environment and Heritage.

Sample collection Ninety-four L. kintorei samples were collected from six locations throughout the distribution of the species (Fig 1): Australian Wildlife Conservancy’s Newhaven Wildlife Sanctuary (hereaf- ter, Newhaven; 22° 49' 41'' S, 131° 6' 58'' E; n=30) and Sangster’s Bore in the northeast of their distribution (20° 49' 60'' S, 130° 19' 60'' E; n=23), Petalu-Docker River toward the centre of their distribution (hereafter Docker River; 24° 52' 27'' S, 129° 05' 01'' E; n=1), Uluru-Kata- Tjuta National Park to the east (hereafter Uluru; 25° 18' 44'' S, 131° 01' 07'' E; n=30), Warbur- ton to the west within the Ngaanyatjarra IPA (26° 08' 00'' S, 126° 35' 00'' E; n=2) and Watarru within the IPA of the APY Lands at the southern extent of the distribution (27° 11' 43" S, 129° 54' 48" E; n=8). Samples from Warburton were provided by the Western Australian Museum. Tissue was obtained via tail-tip biopsy and preserved in 90% ethanol. Liopholis kintorei is a species that exhibits kin-based social living, and high natal philopatry [21]. To ensure that

Fig 1. Distribution (shaded) and sampling localities of L. kintorei. Sangster’s Bore (SB), Newhaven (NH), Uluru (U), Docker River (DR), Watarru (WT) and Warburton (WB). Map derived from data [25] under a CC BY license, with permission from the Commonwealth of Australia, original copyright, 2015. doi:10.1371/journal.pone.0128874.g001

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 3 / 14 Regional Genetic Structure of a Threatened Lizard

individuals included in our analyses were not highly related, we included only adults captured from separate, distinct burrow clusters.

Laboratory procedures Whole genomic DNA was extracted from tissue using a modified salting-out protocol [26]. For each sample the mitochondrial gene, NADH dehydrogenase subunit 4 (ND4), was targeted be- cause previous work on reptiles has shown useful levels of variation for intraspecific studies, though its mutation rate is slow enough to allow inference of deeper divergence due to long- term isolation [12,27]. In addition, individuals were genotyped at ten microsatellite loci to characterise fine-scale population dynamics and more recent divergence. Polymerase chain reactions (PCR) for all markers were carried out using a PTC-100 Ther- mocycler (MJ Research, Inc.). Mitochondrial ND4 was amplified using previously developed primers [28]. PCRs were carried out in 20 μL volumes containing 50–100 ng of DNA, 4 uL 5x

GoTaq Flexi Buffer (Promega), 2 mM MgCl2, 0.2 μM of each dNTP, 0.125 μM of each primer (ND4 and tRNA-leu) and 1 U Taq Polymerase (Promega). Thermocycling began with an initial denaturation for 5 min at 94°C, followed by four touchdown cycles with 94°C denaturation for 30 sec, annealing temperatures (55°C, 53°C, 51°C, 49°C) for 30 sec, and 72°C extension for 45 sec. An additional 35 cycles were carried out at an annealing temperature of 47°C, followed by a final 72°C extension step for 10 min. Microsatellite PCRs were carried out in 10 μL volumes containing ~50 ng of DNA. A -29 M13 sequence was added to the 5’ end of each forward primer to allow for the incorporation of a complimentary M13 fluorescent-labelled tag, following the protocol of Schuelke [29]. Four tetranucleotide microsatellite loci were developed concurrent to this study (BX6, CKD, FQR, J3F; see S1 Text for microsatellite design). In addition to these, we utilised six previously devel- oped markers (Est1, Est2, Est9, Est12 [30]; Ecu2, Ecu3 [31]). All microsatellite loci were ampli-

fied with identical reaction conditions: 2 uL 5x GoTaq Flexi Buffer (Promega), 2.5 mM MgCl2, 0.2 μM of each dNTP, 0.02 μM of forward primer, 0.1 μM reverse primer, 0.1 μM of fluoro-la- belled tag (FAM, VIC, NED, or PET) and 1 U Taq Polymerase (Promega). Thermocycling began with an initial denaturation for 3 min at 94°C, followed by five touchdown cycles with 94°C denaturation for 30 sec, annealing temperatures (60°C, 58°C, 56°C, 54°C, 52°C) for 30 sec, and 72°C extension for 45 s. An additional 35 cycles were carried out at an annealing tem- perature of 50°C, followed by a final 72°C extension step for 10 min. PCR products were visual- ized by electrophoresis on 2% agarose gel. All PCR purification, sequencing and fragment separation was performed by Macrogen (Korea).

Data analysis ND4 sequences were checked by eye and aligned with ClustalW, implemented in MEGA 5.0 [32], and submitted to GenBank (Accession numbers KM035773-KM035789). DNA sequences were then translated into amino acid sequences using the vertebrate mitochondrial code. No premature stop codons were observed, indicating that all sequences are true mitochondrial copies. Haplotype and nucleotide diversities were calculated in DnaSP [33]. A minimum-spanning network of ND4 haplotypes was constructed in TCS 1.21 [34]. Glob-

al and pairwise FST, an analogue of FST [35], were calculated from ND4 haplotypic data in Arlequin v3.5 [36] with 1000 permutations. Microsatellite alleles were visualised and scored using Peak Scanner 1.0 (Applied Biosys- tems). To ensure amplification and scoring consistency, at least 10% of samples at each locus were independently rerun and genotyped. Summary statistics, including exact tests for Hardy- Weinberg equilibrium (HWE) and linkage disequilibrium (LD) were conducted in GenAlEx

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 4 / 14 Regional Genetic Structure of a Threatened Lizard

6.4 [37] and GENEPOP 4.2 [38]. Effective population size (Ne) estimates were calculated utiliz- ing the approximate Bayesian framework implemented in ONeSAMP v1.2 [39]. Due to prohib- itively small sample sizes at Docker River and Warburton, these sampling localities were excluded from population-level analyses.

When calculating FST analogues from highly polymorphic data such as microsatellites, with- in-population variance can often approach the level of the total variance, resulting in very low

FST values even when the populations share no alleles [40,41]. Following Hedrick [40] and Meirmans [41], pairwise fixation index values calculated from microsatellite data (hereafter

F 0ST) were standardised using the program RECODEDATA 0.1 [41]. STRUCTURE v.2.3 [42] analysis was used to assess genotypic clustering and assignment probabilities. We examined values of K =1–8 (double the number of sample sites included in the analysis), with 10 replicate runs for each, 105 MCMC iterations burn-in and 104 main itera- tions. Hubisz et al. [43] developed a new model for STRUCTURE, which allows the use of sam- ple-site information. This is different to the initial models including location priors, in that it adds power to analyses, but can disregard site information when true clustering is uncorrelated with sampling locations. We used the ‘admixture’ model with correlated allele frequencies, and repetitions were run with and without location information. The number of genetic clusters (K) was determined using the ΔK method of Evanno et al. [44]. Discriminant analysis of principal components (DAPC) was used to describe the genetic re- lationship between sampling localities. DAPC is a multivariate analysis that first uses principal components analysis (PCA) to transform data into uncorrelated components. These compo- nents are then analysed using a linear discriminant method, minimising within-group variance while maximising among-group variance [45]. Furthermore, this analysis does not assume HWE and LD, which are often violated when working with natural, small and fragmented pop- ulations [45]. DAPC was carried out in the R package adegenet [46], implemented in R 2.12 (R develop- ment core team 2013; www.r-project.org), with K selected using the find.clusters function and Bayesian Information Criterion (BIC). We also ran DAPC using sample locations as groups (K = 4) to assess the differentiation of our sample sites. PCA was performed in R using the dudi. pca function in the package ade4 [47]. Missing data were replaced with the mean (the origin of the X- and Y-axes, as in Horne et al. [48]). Determining the number of principal components (PCs) to retain as predictors for the discriminant analysis requires a balance between the statis- tical power of more PCs, and the stability of assignments, though there is no strict rule. Retain- ing too many PCs with respect to sample size can result in over-fitting the data. This trade-off can be assessed using the a.score function in the R package adegenet [46]. Analyses were car- ried out retaining a conservative 13 PCs, the optimal number suggested by a.score, given our relatively small dataset.

Results Summary statistics Mitochondrial sequence data. Mitochondrial ND4 sequences of 585 bp were successfully amplified from 72 individuals sampled from the six localities. Sequences contained 23 (3.9%) variable sites, of which 18 (3.1%) were parsimony informative, revealing a total of 17 unique haplotypes (Fig 2). Haplotype and nucleotide diversity over all samples were 0.908 and 0.009, respectively, and the variance for both was < 0.0002 (Table 1). All haplotypes sampled at Uluru were unique to that locality. Newhaven and Sangster’s Bore shared haplotypes with only each other, and Docker River and Warburton both shared haplotypes with Watarru. One

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 5 / 14 Regional Genetic Structure of a Threatened Lizard

Fig 2. Haplotype network constructed from 72 Liopholis kintorei mitochondrial ND4 sequences. Each circle represents a unique haplotype, and the number within indicates its frequency. doi:10.1371/journal.pone.0128874.g002

sample from Warburton was unique to that locality, although small sample sizes may be the reason for this. Microsatellite loci. One of the four remaining sample localities was out of HWE (Newha- ven, P < 0.05; Table 2), which may be due to a Wahlund effect from some spatial structure [49]. Following Holm-Bonferroni sequential correction [50], 10 out of 180 locus x locus tests for LD (45 per sampling locality) were significant, all of which were for different locus pairs. The presence of some LD is unsurprising, given that it can be common in threatened species

Table 1. Sample sizes and diversity indices for Liopholis kintorei captured from four sampling localities.

Locality n (ND4) nh h π n (μsat) Na Ra Pa HO HE Newhaven 23 6 0.76 0.0063 30 10.1 6.3 9 0.765 0.780 Sangster’s Bore 23 7 0.78 0.0077 23 10.8 7.2 18 0.811 0.811 Watarru 8 2 0.54 0.0010 8 5.4 5.2 7 0.771 0.688 Uluru 15 2 0.34 0.0006 30 10.5 6.6 18 0.777 0.791 Warburton 2 2 2 Docker River 1 1 1

Diversity indices for Warburton and Docker River could not be calculated because of small sample sizes. Number of samples (n), number of haplotypes

(nh), haplotype diversity (h) and nucleotide diversity (π) are given for mitochondrial ND4 sequences (total n = 72). The number of samples (n), average number of alleles per locus (Na), allelic richness (Ra), number of private alleles (Pa), observed (HO) and expected (HE) heterozygosities are given over ten microsatellite loci (total n = 94). doi:10.1371/journal.pone.0128874.t001

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 6 / 14 Regional Genetic Structure of a Threatened Lizard

Table 2. Locus by locus results for Hardy-Weinberg Equilibrium tests for L. kintorei sample localities.

Locus Newhaven (n = 30) Sangster’s Bore (n = 23) Watarru (n = 8) Uluru (n = 30) BX6 0.170 -0.040 -0.140 0.120 CKD 0.045 0.004 0.143 0.158 FQR -0.019 -0.073 -0.167 0.046 J3F -0.078 0.010 -0.077 -0.169 EST1 0.014 0.130 -0.217 -0.007 EST2 0.036 0.120 0.056 0.192 EST9 -0.063 -0.062 0.167 0.077 EST12 0.217 -0.081 -0.287 -0.028 ECU2 0.017 0.106 -0.021 -0.068 ECU3 -0.048 0.047 -0.077 -0.052 Overall 0.036 0.025 -0.053 0.035

The inbreeding coefficient (FIS) for each locus is given, as well as over all ten loci. Bold values denote significant FIS (P < 0.05). doi:10.1371/journal.pone.0128874.t002

that are expected to have small effective population sizes [18]. Allelic richness of the four sites

ranged from 5.2–7.2, and the number of private alleles from 7–18. Overall FIS ranged from 0.053–0.036, with these positive values probably reflecting a spatial Wahlund effect [49]. Ne estimates for each locality were: Newhaven (Ne = 60.4, 95% CI = 38.8–119), Sangster’s Bore (Ne = 41.5, 95% CI = 23.4–116.9), Watarru (Ne = 9.3, 95% CI = 5.8–15.9), Uluru (Ne = 21.2, 95% CI = 16.7–27.6). Estimates of Ne are sensitive to sample size, and as such ours should be treated with caution due to small sample sizes. Genetic differentiation between localities. All comparisons of genetic differentiation, ex-

cept one (Newhaven-Sangster’s Bore; PND4 =0.108), were high and significant (Table 3), indi- cating very low connectivity between localities. For ND4, overall FST = 0.50, P < 0.00001. Pairwise population differentiation for both ND4 (FST) and microsatellites (F 0ST) was substan- tial (Table 3; Juke’s Cantor distances between localities are given in Table 4). Newhaven and

Sangster’s Bore were the least differentiated from each other, with low and not-significant FST for ND4, though microsatellite differentiation was relatively high. Uluru was the most differen- tiated from all other localities in all comparisons. STRUCTURE analysis yielded a best-fit value of K=2(S1 Fig) without location informa- tion, and K = 4 with location information. When K = 2 was considered, one cluster comprised the Uluru samples, and the other clumped Newhaven, Sangster’s Bore and Watarru together (S1 Fig). When location information was used, STRUCTURE gave a best-fit value of K=4(Fig 3a), and there was clear delineation of sample sites with high assignment probabilities (Fig 3b). A similar result was found in the DAPC: the lowest BIC value was indeed K=2 (BIC = 124) when no location information was used, however the BIC score for K=4 was only slightly

Table 3. Pairwise genetic differentiation between four Liopholis kintorei sampling localities; generated from 585 bp of the mitochondrial ND4 gene (FST; lower diagonal) and ten microsatellite loci (F 0ST; upper diagonal). All values in bold are significant (PND4 < 0.0001; Pmsat < 0.05). Newhaven Sangster’s Bore Watarru Uluru Newhaven 0.285 0.257 0.302 Sangster’s Bore 0.047 0.380 0.440 Watarru 0.494 0.502 0.263 Uluru 0.627 0.622 0.938 doi:10.1371/journal.pone.0128874.t003

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 7 / 14 Regional Genetic Structure of a Threatened Lizard

Table 4. Range of Jukes-Cantor genetic distances between four Liopholis kintorei sampling localities calculated from ND4. Newhaven Sangster’s Bore Watarru Newhaven Sangster’s Bore 0.000–0.016 Watarru 0.007–0.012 0.005–0.014 Uluru 0.007–0.016 0.009–0.017 0.010–0.014 doi:10.1371/journal.pone.0128874.t004

Fig 3. STRUCTURE and DAPC analyses of Liopholis kintorei individuals captured from four locations across their distribution: Newhaven (NH), Uluru (UL), Watarru (WT) and Sangster’s Bore (SB). (a) ΔK values for each potential number of genetic clusters (K) examined, showing a best-fit value of K = 4. (b) Bar plot showing population assignment of individuals from each sample locality. (c) DAPC scatter plot showing K = 4 genotypic clusters. Each point represents an individual genotype, each sample site being depicted by a different colour and encompassed by a 95% confidence ellipse. Eigenvalue plots represent the amount of genetic variation contained in each discriminant factor. doi:10.1371/journal.pone.0128874.g003

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 8 / 14 Regional Genetic Structure of a Threatened Lizard

higher (BIC = 126; a difference in BIC of 0–2 is considered weak [51]). For K=2, Uluru sam- ples were largely separate from the rest (S1 Fig). However, when samples were grouped by loca- tion in the DAPC, there was considerable genetic separation between all four groups, with the largest separation along the first and second discriminant factors lying between Uluru and the other localities (Fig 3c). The optimal number of principal components retained for the DAPC accounted for 54% of the overall variation, and the first two discriminant factors accounted for 91% of this variation in the discriminant analysis. Only plots for K=4 are given here, but see S1 Fig for plots of K=2.

Discussion We show genetic partitioning among regions containing the Vulnerable lizard Liopholis kin- torei. Each of the localities from which L. kintorei was sampled contained similar levels of ge- netic variation, and individuals from the Uluru-Kata-Tjuta region were most genetically distinctive (Tables 1, 3 and 4). Our estimates of genetic divergence, in addition to environmen- tal differences experienced among regions, indicate that each of these are reservoirs of impor- tant genetic variation and point to the risk of outbreeding depression should interbreeding occur. Outbreeding depression is lowered reproductive fitness in generations subsequent to cross- ing of individuals from genetically differentiated parts of their distribution [16]. A decision tree developed by Frankham et al. [16] assists conservation managers to assess the probability of its occurrence, and thus decide whether populations should be kept separate. Applying this to the data for Liopholis kintorei indicates a risk of outbreeding depression should individuals be translocated. The third part of the tree suggests that if sites have been isolated from each other for 500 years or more, there is a high risk of outbreeding depression and they should remain separated. A crude estimate of divergence times based on a commonly cited mitochondrial cali- bration of 1.3–2% sequence divergence per million years [13,52], suggests that the Uluru line- age may have split from the others between 350 kya and 1.31 million years ago. The level of genetic divergence between sampling localities at ND4 (< 2%; Table 4) was below that found in a closely related species, Liopholis inornata (within-species divergence up to 6.1% [12]); however, L. inornata has a much broader distribution and was sampled across a wider area, that may explain the higher within-species divergence reported. When considering clades of L. inornata sampled across similar geographic scales, the level of differentiation within these two Liopholis species was similar. Furthermore, there are environmental differences between the sites that may contribute to localised adaptation, another factor that flags the possibility of outbreeding depression from translocation according to Frankham et al. [16]. Similarly, Crandall et al. [15] propose evaluat- ing ‘ecological exchangeability’ along with estimates of genetic divergence to decide on the parts of species distributions to be managed separately. Given the latitudinal range over which L. kintorei is distributed, it is not surprising that there are environmental differences across our sampling localities. For example, mean monthly minimum and maximum temperatures are consistently and significantly cooler at Uluru than the northernmost locality of Sangster’s Bore (see S2 Text for climate data and statistical tests). Average annual rainfall at Sangster’s Bore is

479 mm compared with 320 mm at Uluru (t32 = 2.09, P=0.045), though rainfall patterns differ with Uluru receiving more winter rain. Sangster’s Bore experiences more days above 35°C and 40°C than Uluru (respectively, 169.6 vs. 109.3 days 35°C and 52.9 vs. 32.1 days 40°C; cli- ! ! mate data from Australian Bureau of Meteorology, www.bom.gov.au; data not available for Watarru). Given their ectothermic physiology, life history traits in reptiles have been demon- strated to be linked to altitudinal or latitudinal variability in climate [52–54]. As such, the

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 9 / 14 Regional Genetic Structure of a Threatened Lizard

significant climatic differences between our sampling localities are likely to be relevant to L. kintorei, and may have led to adaptive differences in, for example, thermal tolerance or seasonal activity. The habitat in which L. kintorei is found in these areas varies also. The Sangster’s Bore and Newhaven localities occur along and adjacent to palaeodrainage lines, with the species’ pre- ferred habitat in semi-saline spinifex plains dominated by soft spinifex (Triodia pungens) and inland tea tree (Melaleuca glomerata). At Watarru, L. kintorei were found within open mulga woodland or Eremophila and woolybutt (Eragrostis eriopoda) grass shrubland, and at Uluru they occur on sand plains and flat swales dominated by either hard (Triodia basedowi) or soft spinifex (T. pungens and T. schinzii)[21]. Uluru was the most genetically distinct in all analyses, and shared no haplotypes with any other site; Watarru was also highly differentiated from the other areas, but shared haplotypes with the other two southwestern localities (Docker River and Warburton). More sampling at Docker River and Warburton is required to determine the true extent of differentiation among these localities. Sangster’s Bore and Newhaven to the north were highly differentiated based on

the microsatellite data set (F 0ST = 0.285), but were not significantly differentiated at mitochon- drial ND4 (FST = 0.047). This discrepancy, taking into account the different inheritance and mutation rates for these genetic markers [55] might indicate a contemporary barrier to dispers- al but higher levels of historical gene flow. As a result, we recognise three main delineations among localities for conservation management: one to the north (Sangster’s Bore and Newha- ven), one to the southeast (Uluru), and one in the southwest (Watarru, Docker River, Warburton). Given the isolation of localities implied by an apparently patchy distribution [20,22] and the genetic differentiation among localities investigated here, genetic diversity, if lost, may not be replenished by migration. Effective population sizes are substantially lower than actual sizes in wild populations, with the ratio between them often approximating 0.1 [56], and up to 0.5 [57]. Census size estimates for L. kintorei are estimated to be low (<500 at Uluru [20]), and genetic Ne estimates in this study were all low (138–232, though these estimates should be treated with caution due to small sample sizes). The Vulnerable status of this species and low estimated population sizes suggest that genetic diversity and viability may be eroded rapidly over time. The threatening processes attributed to the decline of great desert skinks have not been re- moved, and consequently this is likely to continue. If genetic erosion is allowed to proceed, this can render localised parts of the distribution vulnerable to inbreeding and inbreeding depres- sion [58]. Translocations to bring about a so-called ‘genetic rescue’ have been demonstrated to dramatically reverse the effects of inbreeding depression [17,59]. In the case of L. kintorei, unique haplotypes at some localities and high differentiation estimates suggest that if this need eventuates, parts of the distribution selected for translocations need to be carefully chosen to avoid the risk of outbreeding depression. Knowledge of connectivity combined with landscape management of biological processes is needed to conserve biodiversity [15]. Conservation management for L. kintorei should priori- tise the preservation of suitable habitat, in particular addressing recent and localised changes in fire regimes and predation pressure to reduce the risk of further localised population de- clines and thus erosion of genetic diversity. While further sampling needs to be conducted at Watarru, Docker River and Warburton, the evidence suggests three main delineations for man- agement: (1) Uluru to the southeast, (2) Newhaven and Sangster’s Bore to the north, and (3) Watarru, Docker River and Warburton to the southwest. Uluru in particular should be consid- ered separately for management, and this distinctiveness should be recognised if intervention such as translocation or captive breeding is to be undertaken.

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 10 / 14 Regional Genetic Structure of a Threatened Lizard

Supporting Information S1 Text. Characterisation of four polymorphic microsatellite loci for the great desert skink, Liopholis kintorei. (DOCX) S2 Text. Climate data and analysis for northern and southern regions containing Liopholis kintorei. (DOCX) S1 Fig. STRUCTURE and DAPC analyses of Liopholis kintorei individuals using no loca- tion information. (a) ΔK values for each potential number of genetic clusters (K) examined, showing a best-fit value of K = 2. (b) Bar plot showing population assignment of individuals from each sample locality. (c) DAPC plot showing K=2 genotypic clusters. The y-axis repre- sents the density of individuals along the given discriminant function. One cluster (red) com- prised the Uluru samples, and the other (green) clumped Newhaven, Sangster’s Bore and Watarru together. (TIFF)

Acknowledgments Thank you to A. Beattie, M. Gillings, A. Jung, C. Turnbull and M. Whiting for help with sample collection, and to the Western Australian Museum for providing the samples from Warburton. Many thanks to numerous people from Uluru-Kata Tjuta National Park, particularly N. Tjaka- lyiri, J. Clayton, D. Walkabout and K. Bennison. Thanks to the community of Watarru who as- sisted in the project there, particularly F. Young, T. Mervin and M. Pan. Thanks also to the Yuendumu WoC Rangers for their assistance at Sangster’s Bore and to the Docker River WoC Rangers for their assistance at Petalu. Thankyou to staff at the Australian Wildlife Conservancy (AWC), in particular to J. Schofield and D. Moore for invaluable support onsite, and A. James for guidance. K. Miller, K. Smith and M. Gardner gave assistance with the development of mi- crosatellite primers. P. Momigliano offered statistical advice and comments on the manuscript, and we also thank M. Whiting for constructive comments on the manuscript.

Author Contributions Conceived and designed the experiments: SD SM DGC AJS. Performed the experiments: SD SM AJS. Analyzed the data: SD AJS. Contributed reagents/materials/analysis tools: SD SM DGC AJS. Wrote the paper: SD SM AJS.

References 1. Pianka ER, Vitt L. Lizards: windows to the evolution of diversity. University of California Press, Berke- ley; 2003. 2. Wilson S, Swan G. A Complete Guide to Reptiles of Australia ( 3rd ed.). Reed New Holland, Sydney, Australia; 2010. 3. Byrne M, Yeates DK, Joseph L, Kearney M, Bowler J, Williams MA, et al. Birth of a biome: insights into the assembly and maintenance of the Australian arid zone biota. Mol Ecol. 2008; 17: 4398–4417. doi: 10.1111/j.1365-294X.2008.03899.x PMID: 18761619 4. Barr KR, Lindsay DL, Athrey G, Lance RF, Hayden TJ, Tweddale SA, et al. Population structure in an endangered songbird: maintenance of genetic differentiation despite high vagility and significant popu- lation recovery. Mol Ecol. 2008; 17: 3628–3639. doi: 10.1111/j.1365-294X.2008.03868.x PMID: 18643883 5. Gauffre B, Estoup A, Bretagnolle V, Cosson JF. Spatial genetic structure of a small rodent in a heterog- enous landscape. Mol Ecol. 2008; 17: 4619–4629. doi: 10.1111/j.1365-294X.2008.03950.x PMID: 19140985

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 11 / 14 Regional Genetic Structure of a Threatened Lizard

6. Dennison S, Smith SM, Stow AJ. Long-distance gene flow and habitat specificity of the rock- dwelling coppertail skink, Ctenotus taeniolatus. Austral Ecol. 2012; 37: 258–267. 7. Burbidge AA, McKenzie NL. Patterns in the modern decline of western Australia’s vertebrate fauna: causes and conservation implications. Biol Conserv. 1989; 50: 143–198. 8. Morton SR. The impact of European settlement on the vertebrate animals of arid Australia: a conceptual model. Proc Ecol Soc Aust. 1990; 16: 201–213. 9. Read J, Bowen Z. Population dynamics, diet and aspects of the biology of feral cats and foxes in arid South Australia. Wildlife Res. 2001; 28: 195–203. 10. Pianka ER. Habitat specificity, speciation and species density in Australian desert lizards. Ecology. 1969; 50: 498–502. 11. Pianka ER. Zoogeography and speciation of Australian desert lizards: an ecological perspective. Copeia. 1972; 1: 127–145. 12. Chapple DG, Keogh JS, Hutchinson MN. Molecular phylogeography and systematics of the arid-zone members of the Egernia whitii (Lacertilia: Scincidae) species group. Mol Phylogenet Evol. 2004; 33: 549–561. PMID: 15522787 13. Gardner MG, Hugall AF, Donnellan SC, Hutchinson MN, Foster R. Molecular systematics of social skinks: phylogeny and taxonomy of the Egernia group (Reptilia: Scincidae). Zool J Linn Soc. 2008; 154: 781–794. 14. Moritz C. Defining 'Evolutionarily Significant Units' for conservation. Trends Ecol Evol. 1994; 9: 373– 375. doi: 10.1016/0169-5347(94)90057-4 PMID: 21236896 15. Crandall KA, Bininda-Emonds OR, Mace GM, Wayne RK. Considering evolutionary processes in con- servation biology. Trends Ecol Evol. 2000; 15: 290–295. PMID: 10856956 16. Frankham R, Ballou JD, Eldridge MD, Lacy RC, Ralls K, Dudash MR, et al. Predicting the probability of outbreeding depression. Conserv Biol. 2011; 25: 465–75. doi: 10.1111/j.1523-1739.2011.01662.x PMID: 21486369 17. Weeks AR, Sgro CM, Young AG, Frankham R, Mitchell NJ, Miller KA, et al. Assessing the benefits and risks of translocations in changing environments: a genetic perspective. Evol Appl. 2011; 4: 709–725. PMID: 22287981 18. Frankham R, Ballou JD, Briscoe DA. Introduction to Conservation Genetics ( 2nd ed). Cambridge Uni- versity Press, Cambridge, UK; 2010. 19. IUCN. IUCN Red List of Threatened Species. 2014; Version 2014.1. Available: http://www.iucnredlist.org. 20. McAlpin S. A recovery plan for the great desert skink (Egernia kintorei) 2001–2011. Arid Lands Envi- ronment Centre, Alice Springs, Australia; 2001. 21. McAlpin S. Social structure and mating system of the great desert skink, Liopholis kintorei. Masters Thesis, Macquarie University. 2011. 22. McAlpin S, Duckett P, Stow A. Lizards cooperatively tunnel to construct a long-term home for family members. PLOS ONE. 2011; 6(5): e19041. doi: 10.1371/journal.pone.0019041 PMID: 21589923 23. Pearson D, Davies P, Carnegie N, Ward J. The great desert skink (Egernia kintorei) in Western Austra- lia: distribution, reproduction and ethno-zoological observations. Herpetofauna. 2001; 31: 64–68. PMID: 11169439 24. Cogger H, Cameron E, Sadlier R, Eggler P. The Action Plan for Australian Reptiles. Australian Nature Conservation Agency, Canberra, Australia. 1993. 25. Department of the Environment. Liopholis kintorei in Species of National Environmental Significance Map Summary Version 2. Department of the Environment, Canberra. 2015. Available: http://www. environment.gov.au/science/erin/databases-maps/snes 26. Sunnucks P, Hales DF. Numerous transposed sequences of mitochondrial cytochrome oxidase I-II in aphids of the genus Sitobion (Hemiptera: Aphididae). Mol Biol Evol. 1996; 13: 510–524. PMID: 8742640 27. Greaves SN, Chapple DG, Gleeson DM, Daugherty CH, Ritchie PE. Phylogeography of the spotted skink (Oligosoma lineoocellatum) and green skink (O. chloronoton) species complex (Lacertilia: Scinci- dae) in New Zealand reveals pre-Pleistocene divergence. Mol Phylogenet Evol. 2007; 45: 729–739. PMID: 17643320 28. Arevalo E, Davis SK, Sites JW. Mitochondrial DNA sequence divergence and phylogenetic relation- ships among eight chromosome races of the Sceloporus grammicus complex (Phrynosomatidae) in central Mexico. Systematic Biol. 1994; 43: 387–418. 29. Schuelke M. An economic method for the fluorescent labeling of PCR fragments. Nat Biotechnol. 2000; 18: 233–234. PMID: 10657137

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 12 / 14 Regional Genetic Structure of a Threatened Lizard

30. Gardner MG, Cooper SJ, Bull CM, Grant WN. Isolation of microsatellite loci from a social lizard, Egernia stokesii, using a modified enrichment procedure. J Hered. 1999; 90: 301–304. 31. Stow AJ. Microsatellite loci from the Cunningham’s skink (Egernia cunninghami). Mol Ecol Notes. 2002; 2: 256–257. 32. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. MEGA5: Molecular evolutionary genet- ics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol. 2011; 28: 2731–2739. doi: 10.1093/molbev/msr121 PMID: 21546353 33. Librado P, Rozas J. DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bio- informatics. 2009; 25: 1451–1452. doi: 10.1093/bioinformatics/btp187 PMID: 19346325 34. Clement M, Posada D, Crandall KA. TCS: A computer program to estimate gene genealogies. Mol Ecol. 2000; 9: 1657–1659. PMID: 11050560 35. Excoffier L, Smouse P, Quattro J. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics. 1992; 131: 479– 491. PMID: 1644282 36. Excoffier L, Lischer HE. Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Res. 2010; 10: 564–567. doi: 10.1111/j.1755-0998. 2010.02847.x PMID: 21565059 37. Peakall R, Smouse PE. GENALEX 6: Genetic analysis in Excel. Population genetic software for teach- ing and research. Mol Ecol Notes. 2006; 6: 288–295. 38. Raymond M, Rousset F. GENEPOP (version 1.2): population genetics software for exact tests and ecu- menicism. J Hered. 1995; 86: 248–249. 39. Tallmon DA, Koyuk A, Luikart G, Beaumont MA. ONESAMP: a program to estimate effective population size using approximate Bayesian computation. Mol Ecol Resour. 2008; 8: 299–301. doi: 10.1111/j. 1471-8286.2007.01997.x PMID: 21585773 40. Hedrick PW. A standardized genetic differentiation measure. Evolution. 2005; 59: 1633–1638. PMID: 16329237 41. Meirmans PG. Using the AMOVA framework to estimate a standardized genetic differentiation mea- sure. Evolution. 2006; 60: 2399–2402. PMID: 17236430 42. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000; 155: 945–959. PMID: 10835412 43. Hubisz MJ, Falush D, Stephens M, Pritchard JK. Inferring weak population structure with the assistance of sample group information. Mol Ecol Resour. 2009; 9: 1322–1332. doi: 10.1111/j.1755-0998.2009. 02591.x PMID: 21564903 44. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Mol Ecol. 2005; 14: 2611–2620. PMID: 15969739 45. Jombart T, Devillard S, Balloux F. Discriminant analysis of principal components: a new method for the analysis of genetically structured populations. BMC Genet. 2010; 11: 94. doi: 10.1186/1471-2156-11- 94 PMID: 20950446 46. Jombart T. adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics. 2008; 24: 1403–1405. doi: 10.1093/bioinformatics/btn129 PMID: 18397895 47. Chessel D, Dufour A, Thioulouse J. The ade4 package—I: One-table methods. R News. 2004; 4: 5–10. 48. Horne JB, Momigliano P, Welch DJ, Newman SJ, Van Herwerden L. Limited ecological population con- nectivity suggests low demands on self-recruitment in a tropical inshore marine fish (Eleutheronema tetradactylum: Polynemidae). Mol Ecol. 2011; 20: 2291–306. doi: 10.1111/j.1365-294X.2011.05097.x PMID: 21518062 49. Wahlund S. Zusammensetzung von populationen und korrelationserscheinungen vom standpunkt der vererbungslehre aus betrachtet. Hereditas. 1928; 11: 65–106. 50. Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat. 1979; 6: 65–70. 51. Raftery AE. Bayesian model selection in social research. Sociol Methodol. 1995; 25: 111–163. 52. Chapple DG, Keogh JK. Parallel adaptive radiations in arid and temperate Australia: molecular phylo- geography and systematics of the Egernia whitii (Lacertilia: Scincidae) species group. Biol J Linn Soc. 2004; 83: 157–173. 53. Forsman A, Shine R. Parallel geographic variation in body shape and reproductive life history within the Australian scincid lizard Lampropholis delicata. Funct Ecol. 1995; 9: 818–828. 54. Sunday JM, Bates AE, Dulvy NK. Global analysis of thermal tolerance and latitude in ectotherms. P Roy Soc B-Biol Sci. 2011; 278: 1823–1830. doi: 10.1098/rspb.2010.1295 PMID: 21106582

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 13 / 14 Regional Genetic Structure of a Threatened Lizard

55. Sunnucks P. Efficient genetic markers for population biology. Trends Ecol Evol. 2000; 15: 199–203. PMID: 10782134 56. Frankham R. Effective population size/adult population size ratios in wildlife: a review. Genet Res. 1995; 66: 95–107. 57. Mace GM, Lande R. Assessing extinction threats: toward a reevaluation of IUCN threatened species categories. Conserv Biol. 1991; 5: 148–157. 58. Stow AJ, Sunnucks P. Inbreeding avoidance in Cunningham’s skinks (Egernia cunninghami) in natural and fragmented habitat. Mol Ecol. 2004; 13: 443–447. PMID: 14717898 59. Madsen T, Stille B, Shine R. Inbreeding depression in an isolated population of adders Vipera berus. Biol Conserv. 1996; 75: 113–118.

PLOS ONE | DOI:10.1371/journal.pone.0128874 June 10, 2015 14 / 14

ANIMAL RESEARCH AUTHORITY (ARA)

AEC Reference No.: 2011/037-6 Date of Expiry: 31 August 2014 Full Approval Duration: 1 September 2011 to 31 August 2014 (36 months)

This ARA remains in force until the Date of Expiry (unless suspended, cancelled or surrendered) and will only be renewed upon receipt of a satisfactory Progress Report before expiry. Other people participating: Principal Investigator: Martin Whiting 0402 752 229 Dr Adam Stow Steve McAlpin 02 6689 5290 Dept of Biological Sciences Michael Gillings 0408 403 260 (02) 9850 8153 / 0452 180 342 Andrew Beattie 0450 578 762 [email protected] Christine Turnbull 0414 590 400 Jonathan Davis 0407 929 154 Associate Investigators: Siobhan Dennison 0423 775 653

In case of emergency, please contact: Australian Wildlife Conservancy, Newhaven Sanctuary Manager, Northern Territory, +61 8 9864 6000

or the Principal Investigator / Associate Investigators named above The above-named are authorised by MACQUARIE UNIVERSITY ANIMAL ETHICS COMMITTEE to conduct the following research:

Title of the project: Evolution and maintenance of sociality in lizards of the Egernia group

Type of animal research and aims of project: 4 -Research (Human or Animal Biology) – This project aims to characterize the social structure and dispersal characteristics of Liopholis kintorei and investigate the evolutionary processes that have led to this extraordinary behavior, utilizing ecological, behavioural and molecular methods.

All procedures must be performed as per the AEC approved protocol, unless stated otherwise by the AEC and/or AWO.

Maximum numbers approved: Year Year Year Supplier/ Species Sex Procedure Total 1 2 3 Source Capture measurements, tail biopsy, PIT tagging, Liopholis Kintorei M/F 120 80 80 280 Field ecological survey, behaviour observations Capture measurements, tail biopsy, PIT tagging, Liopholis Inornata M/F 60 60 60 180 Field ecological survey, behaviour observations TOTAL 460

Location of research: Australian Wildlife Conservancy – Newhaven Sanctuary, Northern Territory.

Amendments since initial approval: 1. Change of experimental design to include collection of lizard scats (Approved 8 September 2011) 2. Addition of new personnel: a. Michael Gillings (Approved 8 September 2011) b. Andrew Beattie (Approved 8 September 2011) c. Christine Turnbull (Approved 8 September 2011) 3. Change to experimental design (Approved 16 February 2012) 4. Addition of Jonathan Davis as volunteer (Exec Approved 11 July 2013, ratified by AEC 18 July 2013) 5. Change of technique and housing to allow filming (Approved 15 August 2013)

Conditions of Approval: 1. Sept 2011: Having Biosafety approval in place - Please email [email protected] to notify them of the low risk dealings. 2. Sept 2011: Please ensure all participants are informed of any risks associated with collection of faecal matter.

All Permits/Licenses (to obtain and use fauna; to conduct research at interstate/overseas locations; to house animals, etc.) must be obtained prior to work commencing, and copies forwarded to the Animal Ethics Secretariat.

Being animal research carried out in accordance with the Code of Practice for a recognised research purpose and in connection with animals (other than exempt animals) that have been obtained from the holder of an animal supplier’s license.

Professor Mark Connor (Chair, Animal Ethics Committee) Approval Date: 15 August 2013

Adapted from Form C (issued under part IV of the Animal Research Act, 1985)

ANIMAL RESEARCH AUTHORITY (ARA)

AEC Reference No.: 2012/037-2 Date of Expiry: 31 August 2014 Full Approval Duration: 1 September 2012 to 31 August 2015 (36 months)

This ARA remains in force until the Date of Expiry (unless suspended, cancelled or surrendered) and will only be renewed upon receipt of a satisfactory Progress Report before expiry.

Principal Investigator: Associate Investigators: Dr Adam Stow Siobhan Dennison 0423 775 653 Dept of Biological Sciences Martin Whiting 0402 752 229 0452 636 800 [email protected]

In case of emergency, please contact:

the Principal Investigator / Associate Investigator named above OR

Animal Welfare Officer 9850 7758 / 0439 497 383, Australian Wildlife Conservancy, Newhaven Sanctuary Manager, Northern Territory, +61 8 9864 6000

The above-named are authorised by MACQUARIE UNIVERSITY ANIMAL ETHICS COMMITTEE to conduct the following research:

Title of the project: Evolution and maintenance of sociality: behavioural tests in a non-model species, the Great Desert Skink (Liopholis kintorei)

Purpose: 7 – Research: Environmental Study Aims: 1. To investigate inbreeding avoidance 2. To investigate social behavior and parental care 3. To investigate dispersal behaviours

Surgical Procedures category: 3 – Minor conscious intervention

All procedures must be performed as per the AEC approved protocol, unless stated otherwise by the AEC and/or AWO.

Maximum numbers approved: Species Strain Age Sex Total Supplier/ Source 27 - Lizards Great desert skink (Liopholis kintorei) All ages M &F 140 Wild Capture TOTAL 140

Location of research: Location Full street address Newhaven Wildlife Sanctuary, NT Pastoral lease NT Portion 2506 from plan S931085

Amendments since initial approval: 1. Change of technique and housing to allow filming (Approved 15 August 2013)

Conditions of Approval: N/A

All Permits/Licenses (to obtain and use fauna; to conduct research at interstate/overseas locations; to house animals, etc.) must be obtained prior to work commencing, and copies forwarded to the Animal Ethics Secretariat.

Being animal research carried out in accordance with the Code of Practice for a recognised research purpose and in connection with animals (other than exempt animals) that have been obtained from the holder of an animal supplier’s license.

Professor Mark Connor (Chair, Animal Ethics Committee) Approval Date: 15 August 2013

Adapted from Form C (issued under part IV of the Animal Research Act, 1985)