(Title of the Thesis)*

Conservation Genetics and Reproduction in Three Australian Marsupial Species

Emily J. Miller 2008

A dissertation presented to the University of New South Wales in fulfillment of requirement for the degree of Doctorate of Philosophy in Biological Sciences

“…on [the two largest] these islands are large numbers of cats, which are creatures of miraculous form, as big as a hare; the Head is similar to [that] of a Civet cat, the forepaws are very short, about a finger long. Whereon they have five small Nails, or small fingers, as an ape’s fore-paw, and the two hind legs are at least half an ell long, they run on the flat of the joint of the leg, so that they are not quick in running. The tail is very long, the same as a Meerkat [lemur]; if they are going to eat they sit on their hind legs and take the food with their fore-paws and eat exactly the same as squirrels or apes do.”

- Description of the first sighting of a tammar wallaby in the Abrolhos Islands, Western Australia by Francisco Pelsaert from the translation of Heeres (1899) (originally published 1648)

“DNA…doesn’t lie.” - Mark D. B. Eldridge

For my Family

Image on front cover: Macropus eugenii (Tammar wallaby), reproduced from John Gould, The Mammals of Australia, 1863.

iii

Declaration of Originality

I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.

Signed: ……………………………………… Date: …………………………….. Emily J. Miller

I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.

Signed: ……………………………………… Date: …………………………….. Emily J. Miller

Authenticity Statement

I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.

Signed: ……………………………………… Date: …………………………….. Emily J. Miller

Preface

This thesis consists of five stand-alone papers (Chapters two to six) that are being prepared for submission to international journals of high standing. Each chapter is therefore self-contained and some repetition occurs. To prevent unnecessary duplication a single reference list is provided at the end of the thesis formatted in the style for Conservation Biology.

This thesis is a compilation of my own work, with guidance from my principal supervisors, Catherine Herbert and Mark Eldridge. The contributions of co-authors are detailed below. Chapter 1 2 3 4 5 6 7 Conception EJM EJM, EJM, EJM, EJM, EJM, EJM CAH, CAH, CAH, CAH, CAH, MDBE MDBE MDBE MDBE RJ Sample/data NA EJM, EJM, EJM, EJM, EJM, NA Collection CAH, CAH, CAH, CAH, CAH, NT, NT, MDBE, MDBE, JC, BM, BM, KM, JC, JN, MC, MC, NT, JN MW, HR, NM PO, TF NN, BJ NM, KM Analysis NA EJM EJM EJM EJM EJM NA Writing EJM EJM EJM EJM EJM EJM EJM EJM – Emily J. Miller; CAH – Catherine A. Herbert; MDBE – Mark D. B. Eldridge; RJ – Robert Johnson; NT – Neil Thomas; BM – Brian MacMahon; MC – Martin Clarke; HR – Howard Robinson; NN – Nicole Noakes; NM – Nicola Marlow; KM – Keith Morris; PO – Peter Orell; BJ – Brent Johnson; JC – James Cook; JN – Jan Nedved; MW – Michelle Wilson; TF – Terry Fletcher.

vii

Abstract

Many Australian marsupial species require active population management to ensure their survival in the wild. Such management should be based on a sound understanding of species biology. This thesis explores how knowledge of reproduction and genetics can be applied to the management of three Australian marsupial species faced with contrasting management scenarios.

The ‘vulnerable’ greater bilby is the sole remaining desert bandicoot in Australia. They are a secretive, solitary species whose mating system is unclear. This research examined temporal changes in genetic diversity within two captive breeding programs utilising different management strategies. Using seven microsatellite loci, this study found the regular translocation of new individuals into the population maintained genetic diversity. Parentage analysis revealed the bilby to have a promiscious mating system. Sires and non-sires could not be distinguished by morphological traits.

The tammar wallaby is a polygynous, solitary species that is threatened on mainland Australia, but overabundant on some offshore islands. The population genetics of tammars from the Abrolhos Islands in Western Australia were examined using nine autosomal and four Y-linked microsatellite loci, and mitochondrial DNA. There was a relationship between island size, population size and genetic diversity. The Abrolhos populations have significantly lower genetic diversity and are more inbred than mainland tammars and all sampled populations were significantly differentiated. The Abrolhos and mainland populations should be treated as separate Management Units.

The eastern grey kangaroo is a gregarious, polygynous species that is often locally overabundant. To determine traits influencing male reproductive success, behavioural, morphological, physiological and genetic data were examined and showed dominance status, body size and testosterone concentrations were important factors. Sires were also significantly more heterozygous and genetically dissimilar to females, than non- sires.

As body condition influences individual fitness, and management decisions; five body condition indices (BCI) calculated from morphological data were validated using

viii serum biochemistry and haematology in two kangaroo populations with contrasting body condition. Blood parameters were found to be more reliable indicators of condition, questioning the credibility of BCIs currently used in management.

These studies demonstrate the importance of reproductive and genetic data in assisting wildlife management, regardless of a species conservation status.

Acknowledgements

A career in biology was unexpected and arose from a friend (Joanna Gurung) insisting I attend a biology lecture conducted by Professor David Briscoe, as it was apparently unlike anything else. Eight years later the journey has been incredible and I have experienced so much personal and professional growth, travelled to amazing places and worked with fantastic people. However, none of this would have been possible without the love, support and encouragement from my supervisors, family, friends and colleagues.

I am indebted to my supervisors Dr Cath Herbert, Dr Mark Eldridge and Professor Des Cooper. I am privileged to have had the opportunity to work with you all. I am grateful to Cath for her friendship and giving me the opportunity to do a PhD and sparking my initial curiosity in the biology of marsupials. Cath’s passion, enthusiasm and sheer brilliance have enabled me to learn and experience incredible opportunities. I solely credit Cath for my introduction to the art of ‘fondling testicles’ – all in the name of science of course! Without Cath this PhD would not exist.

Mark Eldridge inspired my passion for genetics, as well as broadening my interests/hobbies to include other taxa. From the very first time we did laboratory work together and discovered sneaky copulations occurring in tammar wallabies, I have never looked back. Mark has been a pillar of support for me during my PhD through continual encouragement and teaching me to appreciate the eccentricities that come with being a biologist. Mark has provided me very useful insights at times of crisis such as “there is something fundamentally wrong with the universe” and “there is nothing worse than a bit of integer overflow”.

Professor Des Cooper has enlightened me over the years learning the intricacies of cricket, and cricket players and possesses inspiring wisdom and knowledge. This PhD involved much fieldwork. James Cook and Jan Nedved dedicated a large amount of their time and energy into capturing kangaroos and wallabies so I could ‘fondle their testicles’, as James puts it. James provided much support, particularly when dealing with spiders. A big thankyou to all the volunteers who have assisted with fieldwork.

To Lee Ann Rollins, Joachim Elenz, and the Elenz clan (Mischa, Adam and Klara) who have become dear friends and a second family to me. Thank you all for keeping me sane, particularly Lee who always help me keep things in perspective and company in the lab. The Rollins/Elenz family has treated me as one of their own and words cannot express my gratitude and how much you have all come to mean to me.

To my dear friends, Richard (Dr Dick) Lane, Ryan (no name) Moore, (Ranger) Dan Tilbury, without your love, support and shenanigans I would not have stayed “sane”. I know you will share the same excitement and relief upon my completion. I love you all! A big thanks to Paul Herbert for all the stirring and motivation, the nicknames I will never live down and the endless hours devoted to my fish tank to keep me diverted. Maria Cardoso and Mick Freeman who have keep me grounded and provided a good laugh whenever needed. Tiffanie Nelson provided great support and friendship. I think we were separated at birth in a previous life; Michael Whitehead who has changed me forever, always encouraged my warped sense of humour, and nurtured my nerdiness. To my many other friends that were pivotal in making this thesis happen: Enhua Lee, Alex James, Fiona Thomson, Celine Frere, The entire PBSG Clan, Matt Fahey, Illara Clyde, Kellie Wilson, and Boyd the Barman for the best Guinness poured in Sydney.

A special thanks to the Molecular Ecology and Evolutionary Facility Members – Jennifer Sinclair, Michael Whitehead, Celine Frere, Maria Cardoso, Gianluca Maio, Merel Dalebout, Jackie Chan, Steve Hamilton – and the founder Professor Bill Sherwin. Bill has taught me that the world can be explained by mathematical equations; thank you Clare for letting me play with the MEEF technician; Jen and Clare for the useful manuscript comments; the ‘Samuels Crew’ and all those creatures in the dungeon; Kris Carlyon, Snoop Lothian, Susi Zajitschek, Angela Moles, Russell Bonduriansky, Rob Brooks, Jonathon Russell, Nelika Hughes; Angela Higgins and Jeremy Shearman (Ramaciotti Centre, UNSW) for processing my GeneScan and sequencing; David Warton for statistical advice (Chapter 5).

This research would not be possible without the animals and staff at Waratah Park Earth Sanctuary and Australia Walkabout Wildlife Park; or the many wonderful people

xi with whom we have collaborated with from the Department of Environment and Conservation, Western Australia – Keith Morris, Neil Thomas, Nicole Noakes, Brent Johnson, Peter Orell, Nicola Marlow, Brian MacMahon, Martin Clarke, Howard Robinson; the Taronga Zoo staff and vets, in particular Dr Robert Johnson, Dr Larry Vogelnest and Dr Kimberly Vinette-Herin; Mark Adams (SA Museum) for access to the wild WA bilby samples. Karina Acevedo-Whitehouse for access to the IR macro; Cameron Wood and the staff at the Royal North Shore Hospital; Jay Cox and Captain Crankypants (The Rat Patrol) for sailing us around the Abrolhos Islands.

I would like to acknowledge the ARC Kangaroo Genome Centre of Excellence and the Fertility Management of Koalas and Kangaroos Contraception Program who provided funding for this research. Some funding was also provided by the State Trustees M. A. Ingram Trust and The Royal Zoological Society of New South Wales. All experimental work in the Chapter 4 was approved by the Department of Environment and Conservation (WA) Animal Ethics Committee under the approval numbers 10/2005 and 40/2007. All experimental work in Chapters 5 and 6 was carried out was approved by the University of New South Wales Animal Ethics Committee under the approval number 05/121.

Last but not least, a big thank you to my family. Dad, Mum, Ang, Ben, Michelle, Lachlan, Alison, Nat, Hannah, Summer, Nana, Jon, Andrew and Melinda. You have all constantly been there to listen to me and offer advice. Had it not been for you support, I would not have undertaken such a conquest (and finished!). I love you all very much and am grateful that I am part of the family.

xii

Conference Presentations

Miller, E. J., Eldridge, M. D. B., Morris, K, and Herbert, C. A (2008) Swimming tammars? Relationships amongst the tammar wallaby (Macropus eugenii) populations in the Houtman Abrolhos Archipelago, Western Australia. Oral presentation at the Australian Mammal Society Conference, Darwin, Northern Territory, Australia.

Herbert, C. A., Morris, K., Orell, P. Miller, E. J., Eldridge, M. D. B., and Renfree, M. (2008) Living on the edge: reproductive ecology of tammar wallabies inhabiting the Abrolhos Archipelago, Western Australia. Australian Mammal Society Conference, Darwin, Northern Territory, Australia.

Miller, E. J., Eldridge, M. D. B., and Herbert, C. A. (2007) Morphological, socio-endocrine and genetic traits that influence male reproductive success in the eastern grey kangaroo (Macropus giganteus). Oral presentation at the 6th International Zoo and Wildlife Research Conference of Behaviour, Physiology and Genetics, Berlin, Germany

Miller, E. J., Eldridge, M. D. B., Cooper, D. W. and Herbert, C. A. (2007) Sex in marsupials. Oral Presentation at the Australian Museum Research Forum, Sydney, New South Wales, Australia.

Miller, E. J., Eldridge, M. D. B., Herbert, C. A. et al (2007) Bigger is not better: male reproductive success in a captive breeding program for the greater bilby (Macrotis lagotis) Oral presentation at the Genetics Society for Australia Conference, Sydney, New South Wales, Australia.

Miller, E. J., Eldridge, M. D. B., Herbert, C. A. et al (2006) Traits that influence male reproductive success in the greater bilby (Macrotis lagotis): implications for management. Oral presentation at the Australasian Wildlife Management Society Conference, Auckland, New Zealand.

Miller, E. J., Eldridge, M. D. B., Cooper, D. W. and Herbert, C. A. (2006) Morphological and genetic traits potentially influencing male reproductive success in the eastern grey kangaroo (Macropus giganteus): implications for management. Poster presentation at the Australian Mammal Society Conference, Melbourne, Victoria, Australia.

xiii

Table of Contents

Declaration of Originality ...... iii Copyright Statement ...... iv Authenticity Statement...... v Preface...... vi Abstract ...... vii Acknowledgements...... ix Conference Presentations...... xii Table of Contents...... xiii List of Tables ...... xviii List of Figures ...... xxi List of Plates...... xxiv Chapter 1 General Introduction ...... 25 1.1 Conservation Biology...... 25 1.2 Conservation Genetics...... 27 1.3 Mating Systems and Reproduction...... 30 1.4 Wildlife Management...... 32 1.5 Study Species ...... 36 1.5.1 The Greater Bilby (Macrotis lagotis Reid 1837)...... 36 1.5.2 The Tammar Wallaby (Macropus eugenii Desmarest 1817) ...... 37 1.5.3 The Eastern Grey Kangaroo (Macropus giganteus Shaw 1790) ...... 40 1.6 Study Aims ...... 42 Chapter 2 Bilbies behind bars: the impact of captive management on genetic diversity in a threatened species ...... 45 2.1 Introduction ...... 45 2.2 Materials and Methods ...... 49 2.2.1 Study populations ...... 49 2.2.2 Microsatellite genotyping ...... 51 2.2.3 Genetic analyses ...... 52 2.3 Results ...... 54 2.3.1 Genetic diversity...... 54

xiv

2.3.2 Temporal changes in genetic diversity ...... 55 2.3.3 Genetic diversity in comparison to wild populations ...... 58 2.3.4 Genetic diversity in comparison to studbook estimates...... 59 2.4 Discussion ...... 61 2.4.1 Genetic diversity in the captive bilby populations ...... 61 2.4.2 Comparison to wild populations...... 63 2.4.3 Comparison to studbook estimates ...... 64 2.4.4 Implications for management ...... 65 2.4.5 Conclusions...... 66 Chapter 3 The genetic mating system, male reproductive success and selection on male traits in the Greater Bilby (Macrotis lagotis) ...... 68 3.1 Introduction ...... 68 3.2 Materials and Methods ...... 72 3.2.1 Study population, data and sample collection ...... 72 3.2.2 Microsatellite genotyping ...... 74 3.2.3 Parentage analysis...... 74 3.2.4 Male reproductive success and selection analysis...... 76 3.3 Results ...... 77 3.3.1 Parentage assignment...... 77 3.3.2 Morphological traits and male reproductive success...... 78 3.3.3 Selection analysis...... 80 3.4 Discussion ...... 83 3.4.1 Mating system...... 83 3.4.2 Male reproductive success...... 85 3.4.3 Selection on male traits...... 86 3.4.4 Implications for conservation and management ...... 86 3.4.5 Conclusions...... 87 Chapter 4 Swimming tammars? Genetics of three island populations of tammar wallabies (Macropus eugenii) in the Houtman Abrolhos Acrhipelago, Western Australia ..89 4.1 Introduction ...... 89 4.2 Materials and Methods ...... 96

4.2.1 Study populations ...... 96 4.2.2 Sample collection and DNA extraction ...... 97 4.2.3 Microsatellite amplification and screening...... 98 4.2.4 Mitochondrial DNA amplification and screening ...... 98 4.2.5 Estimates of autosomal and Y-linked microsatellite diversity ...... 99 4.2.6 Estimates of mtDNA haplotypic diversity...... 100 4.2.7 Population structure and gene flow ...... 101 4.2.8 Phylogenetic analysis...... 101 4.2.9 Computer simulations...... 102 4.3 Results ...... 103 4.3.1 Autosomal and Y-linked microsatellite diversity ...... 103 4.3.2 mtDNA diversity ...... 106 4.3.3 Population structure and gene flow ...... 110 4.3.4 Phylogenetics...... 111 4.3.5 Computer simulations...... 113 4.4 Discussion ...... 114 4.4.1 Genetic diversity...... 114 4.4.2 Population structure and gene flow ...... 119 4.4.3 North Island founder and population genetics...... 120 4.4.4 Phylogenetics and conservation significance ...... 121 4.4.5 Conclusions...... 123 Chapter 5 Dominance, body size and internal relatedness influence male reproductive success in eastern grey kangaroos (Macropus giganteus)...... 124 5.1 Introduction ...... 124 5.2 Materials and Methods ...... 129 5.2.1 Study populations ...... 129 5.2.2 Animal capture and handling...... 130 5.2.3 Behavioural dominance ...... 132 5.2.4 Genetic assignment of paternity ...... 132 5.2.5 Paternity analysis ...... 133 5.2.6 Testosterone radioimmunoassay...... 133 5.2.7 Data analysis...... 134

xvi

5.3 Results ...... 136 5.3.1 Male dominance...... 136 5.3.2 Characterisation of microsatellite loci and paternity assignment ...... 139 5.3.3 Traits that influence male reproductive success ...... 139 5.3.4 Relatedness, heterozygosity and reproductive success...... 141 5.4 Discussion ...... 142 5.4.1 Male dominance...... 142 5.4.2 Male reproductive success, dominance, body size and testosterone ...... 143 5.4.3 Male reproductive success and genetic traits...... 144 5.4.4 Implications for management ...... 146 5.4.5 Conclusions...... 147 Chapter 6 Validation of body condition indices using serum biochemistry and haematology in a Macropodid species ...... 148 6.1 Introduction ...... 148 6.2 Materials and Methods ...... 151 6.2.1 Study populations ...... 151 6.2.2 Sample collection...... 152 6.2.3 Biochemical and haematological analyses ...... 154 6.2.4 Kidney fat index ...... 155 6.2.5 Data Analysis...... 156 6.3 Results ...... 157 6.3.1 Serum biochemistry and haematology...... 158 6.3.2 Validation of assumptions for OLS ...... 161 6.3.3 Body condition indices ...... 164 6.3.4 Validation of body condition indices using blood parameters ...... 167 6.4 Discussion ...... 167 6.4.1 Serum biochemistry and haematology...... 168 6.4.2 Body condition indices ...... 170 6.4.3 Relationship between blood parameters and body condition indices...... 170 6.4.4 Conclusions...... 172

xvii

Chapter 7 General Discussion: Marsupial genetics, reproduction and the implications for wildlife management ...... 173 References ...... 182 Appendix 2 ...... 218 Appendix 3 ...... 220 Appendix 4 ...... 223 Appendix 5 ...... 225 Appendix 6 ...... 227 Appendix 7 ...... 228 Appendix 8 ...... 229 Appendix 9 ...... 232 Photo Gallery...... 236

xviii

List of Tables

Table 1-1 Comparison of the biology of the three Australian marsupials species selected for this study, the greater bilby (Macrotis lagotis), tammar wallaby (Macropus eugenii) and eastern grey kangaroo (M. giganteus)...... 44 Table 2-1 Summary of the overall genetic diversity calculated across all years from seven microsatellite loci for two captive bred populations of the greater bilby (Macrotis lagotis), Return to Dryandra (Dryandra) and Peron Captive Breeding Centre (Peron), Western Australia (mean ± se)...... 54

Table 2-2 Pair-wise genetic differentiation (FST) partitioned over time for two captive bred populations of greater bilbies, Macrotis lagotis, (a) Return to Dryandra (Dryandra) and (b) Peron Captive Breeding Centre (Peron), Western Australia...... 58 Table 2-3 Genetic diversity for wild Northern Territory and Queensland populations, in comparison to the captive Western Australian (WA) populations of greater bilbies (Macrotis lagotis)...... 59 Table 2-4 Comparison of the genetic diversity in the captive greater bilby (Macrotis agotis) populations with other wild marsupial populations of varying conservation status...... 64 Table 3-1 Male morphological traits (mean ± se) measured in the greater bilby (Macrotis lagotis) to examine the strength of selection...... 81 Table 3-2 Standardised linear gradients () and matrix of quadratic and correlational selection gradients () in the greater bilby (Macrotis lagotis)...... 81 Table 3-3 The M matrix of eigenvalues from the canonical analysis of presented in Table 3-2...... 82 Table 4-1 Summary of autosomal microsatellite diversity indices for the East Wallabi Island (EWI), West Wallabi Island (WWI) and North Island (NI) tammar wallaby (Macropus eugenii) populations, and compared to a mainland Western Australian population...... 104

Table 4-2 Genetic differentiation (FST) amongst East Wallabi Island (EWI), West Wallabi Island (WWI) and North Island (NI) tammar wallaby (Macropus eugenii) populations, and the mainland Western Australian population...... 104

xix

Table 4-3 Characteristics of the four Y-linked microsatellite loci in the Abrolhos tammar wallaby (Macropus eugenii) populations (East Wallabi Island, EWI n = 20; West Wallabi Island, WWI n = 16, North Island, NI n = 19)...... 105 Table 4-4 Summary of mtDNA control region (642 bp) diversity indices for East Wallabi Island (EWI), West Wallabi Island (WWI) and North Island (NI) tammar wallaby (Macropus eugenii) populations...... 107 Table 5-1 Number of eastern grey kangaroos (Macropus giganteus) observed (sampled and genotyped) in each of the three semi-free ranging populations...... 130 Table 5-2 Summary of the dominance rank (I) for each male, and hierarchy linearity (h) for three semi-free ranging captive populations of eastern grey kangaroos (Macropus giganteus)...... 137 Table 5-3 Summary of the size range of the 10 morphological traits measured in male eastern grey kangaroos (Macropus giganteus) across all individuals and the mean (± se) for alpha and lower ranked males...... 138 Table 5-4 Summary of the mean allelic diversity (AD), range of alleles, mean and

range of observed (Ho) and expected (He) heterozygosity for the three populations of eastern grey kangaroos (Macropus giganteus) used in this study...... 139 Table 6-1 Summary of the body condition indices used to compare two populations of eastern grey kangaroos (Macropus giganteus) representing a population in ‘good’ condition and a population in ‘poor’ condition....154 Table 6-2 Summary of the main causes of fluctuation in the selected serum biochemistry and haematology parameters used in this study. The arrows denote an increase () or decrease () in the circulating levels of analytes ...... 155 Table 6-3 Summary of the mean (± se) and reference range of the serum biochemistry and haematology parameters for two eastern grey kangaroo (Macropus giganteus) populations. Population A = ‘good’ condition, and Population B = ‘poor’ condition in (a) females, and (b) males. ‘Pooled’ represents the mean calculated from pooled population data from A and B

and therefore depicts the mean values across the full spectrum of body conditions in this species...... 159 Table 6-4 Summary of morphological measurements (mean ± se) for females and males in two populations of eastern grey kangaroos (Macropus giganteus) (Population A = ‘good’ condition, and Population B = ‘poor’ condition). * denotes a significant difference between populations...... 161

xxi

List of Figures

Figure 2-1 Distribution of the greater bilby (Macrotis lagotis) in Australia (black = present distribution; mid-grey = historic distribution at European settlement; pale grey = Late-Holocene sub-fossil). The locations of the two captive breeding programs referred to in this study are indicated by black circles...... 48 Figure 2-2 Genetic diversity over time for the greater bilby (Macrotis lagotis), captive breeding programs, Return to Dryandra (Dryandra, ) and Peron Captive Breeding Centre (Peron, ), Western Australia. (a) Mean allelic

diversity (AD) and (b) mean heterozygosity (He)...... 56

Figure 2-3 Inbreeding coefficients (FIS) over time for two captive bred populations of greater bilbies (Macrotis lagotis); Return to Dryandra (Dryandra, ) and Peron Captive Breeding Centre (Peron, ), Western Australia. The dashed arrows indicate when new individuals were translocated into each colony (Dryandra below graph, Peron above graph) and the number represents the number of individuals...... 57 Figure 2-4 Comparison of the studbook (dashed line) and genotypic (solid line) estimates of (a) genetic diversity, and (b) inbreeding coefficients between 1999 and 2007 for the Peron Captive Breeding Centre (Peron) colony, Western Australia...... 60 Figure 3-1 Percent (%) males siring offspring (± se) between 2000 and 2004 in a semi free-ranging captive greater bilby (Macrotis lagotis) population. ..79 Figure 3-2 Relationship between male body weight (g) and the number of offspring sired between 2000 and 2004 in a semi free-ranging captive greater bilby (Macrotis lagotis) population...... 79 Figure 3-3 Relationship between body weight (g) of male greater bilbies (Macrotis lagotis) and the females with which they sired offspring in a semi free- ranging captive population...... 80 Figure 3-4 Visualisation of the fitness surface on the two major axes of nonlinear

selection, m1 and m4. Reproductive success of male greater bilbies (Macrotis lagotis) was standardised to a mean of one...... 82 Figure 4-1 Present and former distribution of the tammar wallaby (Macropus eugenii) (black = present distribution; grey = historic; (A) indicates the

xxii

location the Houtman Abrolhos Archipelago and (B) the Tutanning Nature Reserve, Western Australia...... 93 Figure 4-2 Geographic relationship of East Wallabi, West Wallabi, and North Islands, within the Wallabi Group of the Houtman Abrolhos Archipelago, Western Australia (Hesperian 2007)...... 97 Figure 4-3 Distribution of Y-linked microsatellite haplotypes across the three Abrolhos Island tammar wallaby (Macropus eugenii) populations, East Wallabi Island (black), West Wallabi Island (shaded) and North Island (white)...... 105 Figure 4-4 Distribution of mtDNA control region haplotypes across the three Abrolhos Island tammar wallaby (Macropus eugenii) populations, East Wallabi Island (black), West Wallabi Island (shaded) and North Island (white)...... 106 Figure 4-5 Variable sites within the examined 595 base pair segment of mtDNA control region from 102 tammar wallabies (Macropus eugenii)...... 109 Figure 4-6 Posterior probabilities from the STRUCTURE analysis indicated the data were structured into two clusters. (a) The log probability data, Ln P(D), as a function of K; (b) The rate of change in the log probability of data, K, as a function of K...... 110 Figure 4-7 The level of admixture among tammar wallabies (Macropus eugenii) in the three Wallabi Group Islands. Each vertical bar represents a single individual...... 111 Figure 4-8 Maximum likelihood (ML) tree of relationship amongst Abrolhos, mainland WA and SA tammar wallaby (Macropus eugenii) mtDNA control region sequences (595 bp). Robustness is indicated by bootstrap values ( 50%, above branches) and Bayesian posterior probabilities ( 0.50, below branches)...... 112 Figure 4-9 The predicted rate of loss of genetic diversity (allelic diversity, AD and

heterozygosity, He) for (a) East Wallabi Island and (b) West Wallabi Island populations of tammar wallabies (Macropus eugenii) in the

Abrolhos Archipelago (Ne = 40 (); Ne = 60 (); Ne = 100 ())...... 113 Figure 5-1 Relationship between male dominance rank and body weight (kg) in eastern grey kangaroos (Macropus giganteus) in three semi-free ranging

xxiii

captive populations (Population A () n = 6; Population B () n = 9; and Population C () n = 6). Dominance rank (highest rank = 1 to the lowest rank = 9)...... 138 Figure 5-2 Relationship between male dominance rank and serum testosterone concentration (nmol/L) in three semi-free ranging captive populations of eastern grey kangaroos (Macropus giganteus) (a) Population A, (b) Population B, and (c) Population C. Dominance rank (highest rank = 1 to the lowest rank = 9)...... 140 Figure 5-3 Percent (%) offspring sired by each male eastern grey kangaroo (Macropus giganteus) of varying dominance rank in three semi-free ranging captive populations (Population A (black) n = 6; Population B (shaded) n = 9; Population C (white) n = 6). Dominance rank (highest rank = 1 to the lowest rank = 9)...... 141 Figure 5-4 The mean (± se) difference in individual heterozygosity between sires () and non-sires () for internal relatedness (IR), standardised

heterozygosity (SH) and mean d2 in the eastern grey kangaroo (Macropus giganteus)...... 142 Figure 6-1 Graphs of the ln-transformed data showing the relationship between body weight (kg) and leg length (cm) for (a) female (n = 31) and (b) male (n = 20) eastern grey kangaroos (Macropus giganteus) from two populations...... 162 Figure 6-2 Graphs of the pooled population ln-transformed data showing the relationship between body weight (kg) and pes length (cm) for (a) female (n = 31) and (b) male (n = 20) eastern grey kangaroos (Macropus giganteus) from two populations...... 163 Figure 6-3 Individual differences in the relationship between ln-transformed body weight (kg) and leg length (cm) for (a) females, and (b) males in Population A (‘good’ condition; ) and Population B (‘poor’ condition; )...... 165 Figure 6-4 Individual differences in the relationship between ln-transformed body weight (kg) and pes length (cm) for (a) females, and (b) males in Population A (‘good’ condition; ) and Population B (‘poor’ condition; )...... 166

xxiv

List of Plates

Plate 1-1 The greater bilby (Macrotis lagotis)...... 37 Plate 1-2 A female tammar wallaby (Macropus eugenii) with a pouch young...... 40 Plate 1-3 A female eastern grey kangaroo (Macropus giganteus) with a joey...... 42 Plate 2-1 The Dryandra woodlands, Western Australia, that surrounds the Return to Dryandra captive breeding facility and serves as a reintroduction site for the locally extinct greater bilby (Macrotis lagotis)...... 50 Plate 2-2 Predator proof fencing surrounding the Return to Dryandra captive breeding facility for threatened species in Western Australia...... 51 Plate 3-1 Collecting morphological measurements from a male greater bilby (Macrotis lagotis), (a) head length (mm) and (b) testis length (mm)...... 73 Plate 3-2 Inspecting the pouch of a female greater bilby (Macrotis lagotis) that had twins present...... 73 Plate 4-1 The harsh environment on (a) West Wallabi, (b) East Wallabi, and (c) North Islands in the Houtman Abrolhos Archipelago, Western Australia...... 96 Plate 4-2 Evidence of kinked tails in the Abrolhos Island populations of tammar wallabies, Macropus eugenii (a) categorised subjectively from right to left, a major kink, a minor kink and a straight tail, (b) close up of a tail with a more prominent kink...... 116 Plate 4-3 Photographs of sexually mature male tammar wallabies (Macropus eugenii) exhibiting (a) unilateral cryptorchism on North Island, Western Australia, and (b) a male with normal testicles from Kangaroo Island, South Australia...... 117 Plate 5-1 Male eastern grey kangaroos (Macropus giganteus)...... 129 Plate 5-2 Collecting a range of morphological measurements (cm) from male eastern grey kangaroos (Macropus giganteus). (a) head length, (b) forearm length, (c) leg length, (d) pes length, (e) tail length, and (f) testis size...... 131 Plate 6-1 Collecting a blood sample from the lateral tail vein of a male eastern grey kangaroo (Macropus giganteus)...... 154

Chapter 1 General introduction 25

Chapter 1 General Introduction

Many wild animal populations currently require some form of active management to ensure their survival. Human modifications to the environment, including habitat destruction and fragmentation, modification of faunal assemblages through the removal or introduction of predators and competitors, modification of floral communities (e.g. conversion of shrubland to pasture), alteration to fire regimes and water sources, and the process of urbanisation have resulted in dramatic changes to the distribution and abundance of endemic species (Caughley 1987). Since the onset of European settlement in Australia, marsupial populations have fluctuated greatly. In most cases there has been a substantial reduction in the abundance and range of small to medium sized mammals, with many becoming extinct or being eliminated from the mainland (Short & Smith 1994). But, some marsupial populations have benefited and are now considered overabundant at a local or regional scale (Herbert 2007; Kinnear et al. 2002). In some cases these species are secure throughout their range, for example the five large kangaroo species (DEH 2007), while in other cases there are geographic and temporal variations in the species conservation status, for example koalas and some small wallaby species (Herbert 2007). Effective management of species at both ends of this continuum, i.e. from critically endangered to overabundant, requires a thorough understanding of many aspects of species biology.

1.1 Conservation Biology

The goal of conservation biology is to conserve global biodiversity. Conservation biology has been described as a ‘crisis discipline’, and encompasses a broad range of fields including ecology, demography, genetics, wildlife and resource biology (Soulé 1985). The current losses of biodiversity and rates of extinctions are unprecedented (Myers & Knoll 2001) and the role of conservation biology has become crucial in the 21st Century. Of the 5 487 mammals species worldwide, 1 141 (21%) are threatened and since 1600, 76 species of mammal have gone extinct worldwide (IUCN 2008). At

Chapter 1 General introduction 26 least half of these extinctions have occurred in Australia in the last 200 years following European settlement of the continent (IUCN 2008; Short & Smith 1994), and approximately 22% of all Australian species are now in decline (Ceballos & Ehrlich 2002). Deterministic and stochastic threats have been identified as the two main types of threats contributing to extinctions (Caughley 1994). Deterministic threats include habitat loss, introduced species, overexploitation, pollution, disease and climate change (Vié et al. 2008). Stochastic threats are random changes in demographic, environmental, catastrophic and genetic factors (Frankham et al. 2002; IUCN 2006). For many species it is still difficult to quantify threatening processes as their biology is poorly known.

There is evidence that conservation efforts can bring species back from the brink of extinction (Brooke et al. 2008). The conservation status of the black-footed ferret (Mustela nigripes) changed from ‘Extinct in the Wild’ to ‘Endangered’ after a successful reintroduction by the United States Fish and Wildlife Service into eight western states and Mexico between 1991 and 2008 (Vié et al. 2008). Similarly, after successful reintroductions in Mongolia since the early 1990s, the Prezwalski’s horse (Equus prezwalskii) has been downgraded from ‘Extinct in the Wild’ in 1996 to ‘Critically Endangered’ (Vié et al. 2008). Within Australia, several species have had their conservation status downgraded due to intensive conservation efforts including the western quoll (Dasyurus geoffroii), Lumholtz’s tree kangaroo (Dendrolagus lumholtzi), tammar wallaby (Macropus eugenii), western brush wallaby (M. irma), burrowing bettong (Bettongia lesueur), Shark-Bay mouse (Pseudomys fieldi) and the greater stick-nest rat (Leporillus conditor) (IUCN 2008).

Threatened species are typically the focus of conservation efforts and taxa that are common receive little attention (Gaston & Fuller 2007). Common species are fundamental to ecosystem structure and function (Hobbs & Mooney 1998) and since their distribution is typically widespread, their impacts often cover large geographical areas (Gaston 2000). Modifications to the landscape have led to many common species suffering considerable reductions in population size (Hobbs & Mooney 1998), but without the threat of impending extinction (Gaston & Fuller 2007). Other common species have adapted to and exploited the modifications, increasing their distribution

Chapter 1 General introduction 27 and/or abundance within their range (Calaby & Grigg 1989; Caughley 1987). Overabundant or expanding populations can have a number of negative impacts on other native species that are rarer and/or less adaptable (Garrott et al. 1993), including reducing natural diversity by monopolising resources, introducing or spreading infectious diseases and parasites, changing the species composition or relative abundance of sympatric species, and even causing local extinctions (Garrott et al. 1993; Noss 1990; Temple 1990). Therefore managing overabundant species is important for the conservation of threatened species.

1.2 Conservation Genetics

Genetics has become an intrinsic component of conservation biology (Frankel 1974), and an instrumental tool in wildlife management and conservation (Hedrick 2001) for both wild and captive populations. Genetics was applied more generally to conservation in the early 1980s after the establishment of the genetic principles for conservation biology (Frankel & Soulé 1981; Schonewald-Cox et al. 1983; Soulé & Wilcox 1980). Subsequently conservation genetics emerged as a prominent discipline and is defined as

“the application of genetics to preserve species as dynamic entities capable of coping with environmental change. It encompasses genetic management of small populations, resolution of taxonomic uncertainties, defining management units within species and the use of molecular analyses in forensics and understanding species’ biology” (Frankham et al. 2002, pp 1).

Techniques in molecular biology have advanced rapidly during the past few decades (Sunnucks 2000). The first method for measuring genetic diversity was discovered in 1966 using protein electrophoresis (Harris 1966; Lewontin & Hubby 1966). Subsequently, several techniques have been devised for examining genetic diversity, such as DNA fingerprinting (Jeffreys et al. 1985a, b), microsatellites (Jarne & Logoda 1996), and single nucleotide polymorphisms (Brooks 2003). The development of the polymerase chain reaction (PCR) revolutionised genetics as it allowed the

Chapter 1 General introduction 28 amplification of small amounts of template DNA, broadening the scope of questions able to be addressed in conservation biology (Bartlett & Stirling 2003). Continual advances in technology and the development of powerful analysis methods (Luikart & England 1999) has led to the sequencing of mitochondrial and nuclear genes, as well as the recent advent of sequencing entire genomes (Sunnucks 2000). These advances are allowing scientists to tap into data that would be otherwise unobtainable (Sunnucks 2000).

The IUCN Red List has recognised that the threat status of 15% of mammalian taxa could not be determined as they are ‘data deficient’ (IUCN 2008). Conservation genetics has been widely applied in wildlife management as it provides unique insights into species biology and makes available information for prioritising conservation efforts. Conservation genetics is applicable at three levels: that of the species, population and the individual (Frankham 1995a), and all levels are important to conserve (Frankham et al. 2002) as they are fundamental units in ecology and evolution. On a species level, genetics enables the clarification and identification of taxonomic uncertainties, which is fundamental to conserving a species. For example, until recently, only one species of tuatara (Sphenodon genus) was recognised in New Zealand, which represented the sole extant genus of the entire reptilian Order. Molecular analyses have revealed this species should be divided into three distinct taxonomic groups and has had significant implications for conservation planning (Daugherty et al. 1990). The increased sensitivity of technology and methodology has facilitated the detection of introgression (Mallet 2005; Rhymer & Simberloff 1996), which has conservation implications through the loss of the genetic integrity of a species, for example recovery efforts for the highly threatened Ethiopian wolf (Canis simiensis) have struggled to find pure bred animals for captive breeding due to introgression with domestic dogs (Gottelli et al. 1994). Phylogeography has also become an increasingly valuable technique for the identification of a species origin, patterns of geographic dispersal and subsequent colonisation(s) (Edwards et al. 2007). For example, Y-linked microsatellite haplotypes have been used to elucidate the global dispersal patterns of human populations (Kayser et al. 2001).

Chapter 1 General introduction 29

The determination of genetic relationships on a population level is essential for obtaining information on the genetic distinctness of populations, distinct lineages, identification of separate management units, populations of high conservation priority, delineating species mating systems (Frankham et al. 2002; Hughes 1998), as well as the demographic and historical structure of a population (Escudero et al. 2003). For example, the origin of Ireland’s fauna has been the subject of much debate but with the use of mitochondrial DNA, Finnegan et al. (2008) were able to confirm the Irish red squirrel (Sciurus vulgaris) population arose from a colonisation event on the island, as well as delineate regional genetic structure of the species and identify management units for conservation. Genetics is also important for obtaining information on the impact of genetic stochasticity and its effect on random genetic drift, inbreeding, changes in allele frequencies and genetic diversity (Frankham 1995a; Shaffer 1981). Such information is required for the genetic management of species, particularly threatened, captive bred populations to avoid any further genetic deterioration in addition to what may have already occurred in the declining wild population.

The identification of ‘units’ for conservation has become increasingly popular in conservation biology, but also hotly debated (Crandall et al. 2000; Fraser & Bernatchez 2001; Moritz 1994; Moritz 1999; Ryder 1986; Waples 1991), however all aim to conserve the adaptive genetic variance within species (Fraser & Bernatchez 2001). Two prominent concepts for units of conservation are Evolutionary Significant Units (ESUs) and Management Units (MUs) within ESUs (Hedrick et al. 2001c). An ESU is a group of individuals with a similar genetic composition that demonstrate deep evolutionary divergence from other groups within the same species (Frankham et al. 2002; Moritz 1994). The original definitions of ESUs incorporated two main elements: reproductive isolation (i.e. genetic distinctness) and ecological distinctness (Ryder 1986; Waples 1991). More recently, Moritz (1994) shifted the emphasis to genetic distinctness, suggesting ESUs should be recognised as ‘reciprocally monophyletic for mtDNA alleles and show significant divergence of allele frequencies at nuclear loci’ (Moritz 1994). MUs focus on identifying interactions between closely related populations within a species (Moritz 1994; Moritz 1999) and represent the ecological component of ESUs. Populations that have diverged in allele frequency but do not yet

Chapter 1 General introduction 30 show reciprocal monophyly for mtDNA are also significant for conservation, and are referred to as MUs (Moritz 1994; Moritz 1999).

Much research focuses on the individual level, examining individual fitness traits, identifying source populations for individuals, paternity assignment, reproductive success, and mate choice (Charmantier & Sheldon 2006; Mays & Hill 2004). Identifying source populations of individuals is essential for their translocation or reintroduction into other populations to avoid inbreeding and introgression (Frankham 1995a; Frankham et al. 2002). Accurate parentage data is essential for wildlife managers when making decisions regarding the pairing of individuals for mating and the transfer of individuals among wild and captive populations. The widespread application of genetics has also revealed inconsistencies in the presumed relationship between social organisation, parentage and mating system (Ambs et al. 1999; Coltman et al. 1999a; Issac 2005; Worthington Wilmer et al. 1999), transforming how mating systems are understood.

1.3 Mating Systems and Reproduction

Crucial for effective management is a knowledge of a species mating system and parentage to determine effective population sizes, detect inbreeding, genetic differentiation, and genetic diversity, all of which affect population growth, survival and evolutionary potential (Ralls & Ballou 1986; Reed & Frankham 2003). Parentage cannot be determined using behavioural data alone as genetic analyses have revealed inconsistencies in the presumed relationship based on observational data (Ambs et al. 1999; Coltman et al. 1999a; Worthington Wilmer et al. 1999). Mating systems describe the social and genetic relationships of individuals in relation to reproduction (Emlen & Oring 1977). The diversity in mammalian mating systems is a product of the reproductive strategies of individuals and their environments (Clutton-Brock 1989; Emlen & Oring 1977; Reynolds 1996). The classic mating system classifications are monogamy, polygyny, polyandry and promiscuity, though the precise definition of each can differ even within the field of evolutionary biology (Andersson 1994).

Chapter 1 General introduction 31

Typically, monogamy is defined as the continuing bond and exclusive relationship between a male and a female. Polygyny occurs when a single male mates with several females, and polyandry is the reverse, whereby a single female mates with several males (Wittenberger 1981). In a promiscuous mating system, both males and females mate with more than one partner and there is no long term relationship (Wittenberger 1979). Polygyny commonly occurs in species where males do not provide parental care (i.e. most mammals). As a result, males are able to invest their energy into competing for resources and mates (Emlen & Oring 1977). The ability to monopolise resources is a key factor in the intensity of sexual selection. Polygyny is predicted to occur when there is spatial clumping of females in such a way that enables males to defend multiple females from being accessed by other males (Clutton-Brock 1989; Emlen & Oring 1977) and is characterised by greater variance in male reproductive success than in females (Hoogland & Foltz 1982). Polyandrous mating systems are characterised by females pairing with multiple males during a single breeding season, however, true polyandry has been documented in very few species (Wittenberger 1979). In contrast, promiscuity generally occurs when males are unable to successfully monopolise access to females. It often occurs in species where female groups are unstable, males do not provide parental care, and male home ranges are distributed throughout the home ranges of several females (Clutton-Brock 1989). Males are often unable to defend territories and therefore cannot monopolise access to females (Ramsay & Stirling 1986).

According to sexual selection theory (Darwin 1859), males ought to compete for access to receptive females, the most limiting resource for male reproductive success (Trivers 1972). There is evidence in several taxa demonstrating that a larger body size is advantageous when competing for access to receptive females (Clinchy et al. 2004; Fisher & Cockburn 2005; Fisher & Lara 1999). The male-biased sexual size dimorphism that exists in many mammalian taxa is often attributed to competition between individuals for reproductive opportunities, encouraging the evolution of secondary sexual traits (Andersson 1994; Birkhead 2000). Female preference for particular traits could potentially increase the strength of selection on male traits. For example, if females prefer larger males because their size indicates quality, strength, or

Chapter 1 General introduction 32 fighting ability (Andersson 1994; Bercovitch et al. 2003; Coltman et al. 2001; Schulte- Hostedde & Millar 2002), then we would expect strong selection for body size.

Mating system and parentage studies are important for understanding ecological and evolutionary forces influencing individual fitness, and thus reproductive strategies, and are strongly linked with effective population size (Ne). Ne is defined as:

“the number of individuals that would give rise to the calculated inbreeding coefficient, loss of heterozygosity or variance in allele frequency if they behaved in the manner of an idealised population” (Frankham et al. 2002, pp 189).

In mating systems where there is high reproductive skew and some individuals are excluded from breeding (Nunney 1991, 1993), Ne can become very low (Frankham 1995b; Nunney & Elam 1994). In populations of conservation concern, this is of particular importance as Ne determines the rate of loss of genetic diversity and inbreeding which contributes to the risk of extinction (Frankham 1995b; Frankham et al. 2002).

1.4 Wildlife Management

Many wildlife populations need to be actively managed to stem the loss of global biodiversity and extinctions. Due to limited resources available for conservation and management, the application of genetics is an important tool for prioritising conservation efforts. Effective conservation requires the identification of key threatening processes contributing to a species decline, and the implementation of management practices to modify their effects (Caughley & Gunn 1996). The action taken depends on the processes involved and there is often no single solution. Management actions may include increasing the suitability and area of habitat, captive breeding, reintroductions and/or translocations to artificially improve gene flow and re- establishment of populations, cross-fostering and predator control (Lindenmayer & Burgman 2005). Conservation genetics can contribute to each step of the decision-

Chapter 1 General introduction 33 making process through resolving taxonomic uncertainties so that managers are confident of the status of populations, as well as identifying the relationships among the populations they aim to conserve. Decisions regarding the future of wild and captive populations also rely on knowledge of genetic diversity, levels of inbreeding, and the degree of differentiation among populations (Frankham et al. 2002).

Usually the causes of species decline are multifaceted therefore management strategies need to take a similar approach. For example in Western Australia (WA), predation by the introduced European red fox (Vulpes vulpes) has contributed to the decline of many species. In response, the largest wildlife recovery program ever undertaken in Australia, the Western Shield Program, was established in 1996 by Western Australia’s Department of Environment and Conservation (DEC). The main aims of this program are to maximise the recovery of sustainable populations of threatened species through predator control, and reintroduce threatened species into the former habitats where they have become locally extinct (CALM 1999). The program continues to monitor the recovery of these populations, increase awareness though education and public relations, and makes the best use of new and existing research to enhance its recovery programs (CALM 1999). Captive breeding programs were established for 14 threatened species in WA to increase the number of individuals available for wild reintroductions (Mawson 2004a). In response to predator control, population numbers for many species have increased and species have expanded beyond the areas undergoing predator control (Kinnear et al. 2002), which has been instrumental in the delisting of several species. For example the status of the brush-tailed bettong (B. penicillata) has been changed from ‘Endangered’ to ‘Lower Risk’, and both the western quoll (D. geoffroii) and numbat (Myrmecobius fasciatus) have been de-listed from ‘Endangered’ to ‘Vulnerable’ (Orell 2004). Many species have also been reintroduced into areas where they had become locally extinct (Mawson 2004a).

In contrast to the broad pattern of mammal declines across Australia, some species have experienced an increase in abundance and/or have expanded their range in response to human-induced alterations to the landscape (Hobbs & Mooney 1998). ‘Overabundance’ is a value-laden term, which is usually applied to a population when they exceed the carrying capacity of the environment, have undesired effects on the

Chapter 1 General introduction 34 ecosystem and/or other species, or interfere with humans and livestock (Caughley 1987; Herbert 2007). Animals at such high densities pose a risk to their own welfare through a reduction in available resources which leads to reductions in fecundity, disease resistance, and survival. Often welfare considerations play an important role in management decisions, for example culling animals deemed to be in poor condition (DEH 2007). There are several options for managing overabundant populations including doing nothing, translocation, commercial and non-commercial harvesting, restricting dispersal though fencing, reducing fertility (fertility control) through contraception or sterilisation, and euthanasia (Caughley 1987; Coulson 1998; Herbert 2007). The appropriate management strategy employed is usually governed by socio- economic and political factors (Coulson 1998), particularly when these populations are located in close proximity to nature reserves and urban areas (Adderton Herbert 2004), or involve iconic species, for example the koala (Herbert 2007).

The potential evolutionary consequences of harvesting wildlife have received little attention until recently. There is now a growing body of evidence suggesting that practices such as size-selective harvesting (or trophy hunting) can have negative ecological and evolutionary consequences at a population level, for example changes in selective pressures on morphological traits (Allendorf et al. 2008). As males are often larger than females, size-selective harvesting is usually sex-biased towards males. Coltman et al. (2003) have shown that size-selective harvesting in big-horn rams (Ovis canadensis) has lead to a significant reduction in body weight and horn size of males in one population that has been intensively studied for 30 years. In addition to potential phenotypic changes in males, sex-specific culling can influence population dynamics by reducing female fecundity, disrupting gene flow between populations and altering fine scale genetic structure (Allendorf et al. 2008; Harris et al. 2002). There is debate about the extent to which male sex-biased harvesting could limit population size by reducing female fecundity when trophy males are removed (Fairall 1985), with some scientists suggesting that this is unlikely if young males can fertilise females in the absence of trophy males (Mysterud et al. 2002). But, sex-biased harvesting is thought to have led to reproductive collapse in some species, for example several ungulates (Fairall 1985; Milner-Gulland et al. 2003; Mysterud et al. 2002; Solberg et al. 2002) and elephants (Milner-Gulland & Mace 1991). In a long-term (24 year) study

Chapter 1 General introduction 35 of a red deer (Cervus elaphus) population on the Isle of Rum, removal of hunting pressure led to an increase in the number of breeding females, which was associated with a decrease in the levels of polygyny and a decline in female population genetic structure (Nussey et al. 2006). These examples show the extent to which hunting (or the removal of the pressures exerted by hunting) can influence population genetic structure with long term evolutionary consequences. Unfortunately there are very few long-term studies on harvested mammalian populations, so there is very little empirical data on the consequences of selective harvesting. The first step towards developing a greater understanding of the genetic consequences of harvesting is to collect data on the population genetic substructure, mating system and the extent to which targeted traits are heritable in the target populations (Allendorf et al. 2008).

Fertility control is becoming increasingly attractive to control overabundant mammals, particularly highly valued native mammals (Garrott et al. 1993), as it is advocated as a more humane, ethical method than culling. Methods of fertility control include surgical sterilisation, steroidal and non-steroidal contraception, immunocontraception and dopamine agonists that inhibit lactation. The advantages and disadvantages of each methods are reviewed elsewhere (Adderton Herbert 2004; Cooper & Herbert 2001) and will not be dealt with in this thesis. The management of overabundant native species requires a balance between overcoming the negative impacts of the species, yet maintaining an adequate population base with sufficient genetic diversity to ensure species survival within the area. From a genetic perspective, the use of a reversible fertility control is beneficial as it would enable genetic diversity to be maintained in populations (Adderton Herbert 2004; Herbert 2007). This may be an especially important issue for some of the smaller macropodid species that are threatened on a national scale, but have temporal or localised population eruptions such as the black footed rock wallabies (Petrogale lateralis) in WA and koalas (P. cinereus) in Victoria (Herbert 2007). A recent study by Tanaka et al. (submitted) has concluded that regular contraception, in which all females are allowed to reproduce at some stage during their life, will not affect effective population size (Ne in the genetic sense).

Chapter 1 General introduction 36

1.5 Study Species

In order to explore these issues in the context of wildlife management, reproductive and genetic data was collected and applied to three Australian marsupial species: the greater bilby (Macrotis lagotis), tammar wallaby (Macropus eugenii) and the eastern grey kangaroo (M. giganteus). Each species has a different life history strategy and has been affected by anthropogenic changes to the landscape in varying ways, resulting in contrasting current management objectives. The differences between the three study species examined in this thesis are summarised in Table 1-1.

1.5.1 The Greater Bilby (Macrotis lagotis Reid 1837)

The greater bilby is an iconic species for conservation and the sole remaining species of desert bandicoot in Australia (Johnson 2002). Bilbies are distinguished by their soft, silky hair, long rabbit-like ears, and long furry black and white tail (Plate 1-1). They are an omnivorous, nocturnal marsupial that live solitarily or in pairs (Johnson 2002). Bilbies dig burrows up to two metres deep, and individuals may have up to a dozen burrows they use within their home range (Moseby & O'Donnell 2003). Bilbies exhibit sexual dimorphism with males being, on average, one to two times larger than females (males: 800 – 2500g; females: 600 – 1100g). They are capable of breeding all year round, producing one to three offspring per litter, and up to four litters per year under ideal conditions. Young permanently exit the pouch at approximately 75 days, before becoming independent at 90 days of age, and males do not provide parental care (Southgate et al. 2000). The bilby mating system is not well documented due to their secretive nature and the difficulties associated with wild observations. The mating system of the bilby is thought to be polygynous or promiscuous (Johnson & Johnson 1983; Lee & Cockburn 1985; Moritz et al. 1997).

Bilbies have experienced an 80% reduction in their range throughout the arid and semi-arid zones of Australia (Southgate 1990) and are now confined to north-western Central Australia (from the Tanami Desert, south to Warburton, WA,) with some isolated satellite populations in Queensland (QLD) (Southgate 1990). In addition to habitat loss, bilbies have suffered a significant decline in numbers due to competition with introduced species and predation (Moritz et al. 1997), and the species is listed as

Chapter 1 General introduction 37

‘Vulnerable’ nationally (IUCN 2006). As a conservation response, a National Recovery Plan has been established (Pavey 2006) that aims to implement predator control for European red foxes (V. vulpes) and feral cats (Felis catus) in areas where bilbies are in decline, maintain husbandry and captive breeding programs with a coordinated management of populations; reintroduce captive-bred animals to areas of their former range and continue refinement of methodology for monitoring wild populations (Pavey 2006).

Plate 1-1 The greater bilby (Macrotis lagotis).

1.5.2 The Tammar Wallaby (Macropus eugenii Desmarest 1817)

The tammar wallaby is one of the smallest Macropus species, with a dark grey-brown upper coat, a pale buff-grey coat beneath, and reddish arms, feet and flanks (Plate 1-2). Most individuals also have a faint white cheek stripe (Smith & Hinds 2002). Tammar wallabies rest during the day in low, dense vegetation and move into open grassy areas for feeding during dusk and the evening (Smith & Hinds 2002). They are a solitary species whose home range overlaps with several other individuals while they feed in the same area, but no territorial behaviour has been documented (Smith & Hinds

Chapter 1 General introduction 38

2002). The tammar wallaby is sexually dimorphic with males being one to two times larger than females (males: 6 – 10kg; females: 4 – 6kg). Tammar wallabies have a polygynous mating system, in which females usually mate with more than one male (Miller et al. in press; Rudd 1994; Smith & Hinds 2002). Breeding is highly seasonal and synchronised, with a large proportion of births occurring in late January and early February (Tyndale-Biscoe & Renfree 1987). Females exhibit a post-partum oestrus within an hour or two of giving birth and all mating activity usually concludes within four hours of parturition (Renfree et al. 1989; Rudd 1994). The resultant conceptus grows to a 100-cell blastocyst before entering embryonic diapause (Tyndale-Biscoe & Renfree 1987). The loss or removal of a pouch young during the breeding season (the period of decreasing day length from the summer solstice to the winter solstice) stimulates the quiescent blastocyst to reactivate. Birth and a new post-partum oestrus then occurs approximately 26.5 ± 0.4 days later (Renfree et al. 1989). Loss or removal of a young during the non-breeding season (the period of increasing day length between the winter and summer solstice) does not result in reactivation of the quiescent blastocyst (Tyndale-Biscoe & Renfree 1987).

Tammar wallabies have a fragmented distribution with populations inhabiting mainland south-western Australia, Garden Island, two islands in the Recherche Archipelago, and three islands in the Abrolhos Archipelago (WA) and Kangaroo Island, South Australia (SA). The SA mainland, Flinders Island and St Peters Island populations are now listed as ‘Extinct in the Wild’ (Poole et al. 1991). Two subspecies of tammar wallaby are commonly recognised in Australia, M. e. derbianus (WA) and M. e. decres (Kangaroo Island, SA), and despite a growing body of evidence indicating morphological and genetic distinctions, they are generally grouped together and referred to as M. eugenii (DEH 2004). Declines in the tammar wallaby have been attributed to land clearing for agriculture, predation by introduced species and stochastic environmental events such as bushfires (DEH 2004).

The tammar wallaby poses interesting issues for wildlife management. They have suffered a dramatic decline in numbers on the mainland and some islands, but thrive on several other offshore islands. Tammar wallabies are listed as ‘Low Risk (near

Chapter 1 General introduction 39 threatened)’ on a national scale (IUCN 2006), however there is a marked difference between the mainland and island populations. Tammar wallabies from New Zealand that are believed to be descendents from the original mainland SA population (Taylor & Cooper 1999) are currently being bred in captivity for reintroduction to mainland SA (DEH 2004; Taylor & Cooper 1999). On mainland WA, the tammar wallaby was de-listed from the State Threatened Fauna List as a result of their recovery under the Western Shield fox baiting program (Kinnear et al. 2002; Possingham et al. 2003). They are now in sufficiently high densities at a localised scale in the wheat belt of south-western WA that these populations are being used as a source for translocations into other areas of their former range (Mawson 2004b). On several offshore islands they are considered overabundant, for example on Kangaroo Island (SA) tammar wallabies are at such high densities, thousands are culled annually to reduce competition with domestic stock for pasture (Wright & Stott 1999). Despite the tammar wallaby being a model species for a wide range of marsupial research (Hinds et al. 1990; Smith & Hinds 2002; Tyndale-Biscoe & Renfree 1987), surprisingly little is known about their evolutionary ecology.

Chapter 1 General introduction 40

Plate 1-2 A female tammar wallaby (Macropus eugenii) with a pouch young.

1.5.3 The Eastern Grey Kangaroo (Macropus giganteus Shaw 1790)

The eastern grey kangaroo (Plate 1-3) is a crepuscular, polygynous species with a complex social organisation as they are gregarious, territorial and hierarchical (Jarman 1983). Eastern grey kangaroos exhibit pronounced sexual size dimorphism with males being up to four times larger than females at first oestrus (males: 19 – 85kg; females: 13 – 42kg), and males compete intensively for access to resources (Coulson 2008; Ganslosser 1989; Jarman & Southwell 1986). They are capable of breeding all year round, although there is a peak in the summer months, October to March (Poole 2002; Poole & Pilton 1964). This peak in breeding is associated with elevated levels of testosterone concentrations in male kangaroos, but with no significant seasonal variation in testis size (Nave 2002). Females are polyoestrous and monovular, but unlike the majority of species in the family Macropodidae, the gestation period is eight to nine days shorter than the oestrous cycle, and they do not exhibit post-partum

Chapter 1 General introduction 41 oestrus. Instead females return to oestrus 11 days following the loss of pouch young, or during the second half of lactation when the pouch young is greater than 160 days old (Poole & Pilton 1964). Eastern grey kangaroos live in small groups of less than six individuals that consist mainly of mothers and daughters, as young males generally disperse as young adults. At dawn and dusk, several groups whose home-ranges overlap, aggregate where there are concentrated resources, for example a preferred feeding area, forming a mob (Jarman & Southwell 1986).

Eastern grey kangaroos have a wide and almost continuous distribution through the eastern states of Australia, from QLD to Tasmania, from the inland plains to the coast (Poole 2002). They occupy a range of habitats including woodland, shrubland, open forest, and semi-arid mallee and mulga scrubs (Poole 2002) and have increased in number throughout their range in response to landscape modifications such as the conversion of forest and bush land to agriculture and the increase in permanent watering points (Calaby & Grigg 1989; Newsome 1965; Oliver 1986).

Management of kangaroos is a contentious issue in Australia as they are an iconic species. The proximity of many eastern grey kangaroo populations to urban areas makes management operations subject to public scrutiny as they are highly visible (Adderton Herbert 2004). Current management strategies include translocation, fertility control, commercial and non-commercial harvesting, fencing of areas, and euthanasia (Adderton Herbert 2004; Garrott et al. 1993). The formal policy framework for kangaroo management in Australia stipulates that culling should take place to alleviate or prevent the suffering of individuals (DEH 2007). The health and condition of the population is monitored, both prior to the cull to assess its necessity, and after the cull to monitor its success. Commercial harvesting of kangaroos is biased towards the largest males within the population (DEH 2007). Tenhumberg et al. (2004) attempted to model the impact of harvesting on kangaroos and found that size-selective harvesting could result in significantly smaller kangaroos for a given age when the entire population is subject to harvesting. Dispersal of individuals from non-harvested populations into the area may help to mitigate the genetic effects of harvesting, as well as maintain population size and structure (Tenhumberg et al. 2004) which is essential for maintaining their evolutionary potential (Ralls & Ballou 1986; Reed & Frankham

Chapter 1 General introduction 42

2003). But, we currently do not have an understanding of many of the variables necessary for input into these models, such as knowledge of the genetic mating system, the genetic substructure of the population and the heritability of the selected traits.

1.6 Study Aims

This thesis reports research on conservation genetics and reproduction in three Australian marsupial species. The broad aim was to use conservation genetic techniques to explore aspects of the three selected species life history (e.g. mating systems, Chapters 2 and 5), population structure (e.g. delineation of management units, Chapter 4) and the response of populations to management (e.g. captive breeding strategies, Chapter 3). In each case, the data were used to test current conservation genetics theories and to demonstrate how an understanding of population genetics is

Chapter 1 General introduction 43 critical for effective population management. In addition, the final research chapter (Chapter 6) sought to validate five commonly used body condition indices. Such indices are often employed to evaluate population health in the context of management and are an important tool in the field of evolutionary ecology by allowing us to objectively measure the quality of individuals. The specific aims of each chapter were:

(i) To investigate the impact of contrasting management strategies on the temporal patterns of genetic diversity for two captive breeding programs for a threatened species, the greater bilby (Chapter 2). (ii) To elucidate the genetic mating system, patterns of male reproductive success and the strength of sexual selection operating on male morphological traits in an elusive threatened species, the greater bilby (Chapter 3). (iii) To examine the population genetics and patterns of gene flow among three island populations of tammar wallabies and identify whether they are of particular genetic or conservation significance (Chapter 4). (iv) To identify whether particular behavioural, morphological, physiological, or genetic traits influence male reproductive success in the eastern grey kangaroo (Chapter 5). (v) To validate five commonly used body condition indices using serum biochemistry and haematology using the eastern grey kangaroo as a model species (Chapter 6).

Chapter 1 General introduction 44

Species Greater bilby Tammar wallaby Eastern grey kangaroo Conservation Vulnerable Low risk (near threatened) Least concern status

Distribution Once widespread, now confined to Remnant in WA mainland, several offshore Wide almost continuous between inland north-west Central Australia. islands in WA and SA. Extinct on SA and east coast of Australia Satellite populations in QLD mainland and Flinders Island

Habitat Semi-arid to arid Dense vegetation for daytime shelter; open Semi-mallee scrub through to woodland grassy area for feeding to forest; open grassy area for feeding

Active Nocturnal Crepuscular/nocturnal Crepuscular

Social Solitary Live in large mixed sex groups/solitary, not Gregarious, live in large mixed sex organisation territorial, hierarchical groups, territorial, hierarchical

Mating system Uncertain, polygynous or Polygynous Polygynous promiscuous

Sexual Yes Yes Yes - pronounced dimorphism Breeding All year round Highly seasonal, late Jan – early Feb with All year round, births peak in summer secondary peak late Feb – early March (Oct – Mar)

No. offspring 1 -3/litter 1 1

Management Captive breeding and Varied as some populations are threatened Overabundant local populations reintroductions, predator control and others are overabundant. Translocations, throughout range. Several management captive breeding and reintroductions, techniques including culling and fertility predator control and culling control

Chapter 2 Genetic diversity in contrasting captive breeding programs 45

Chapter 2 Bilbies behind bars: the impact of captive management on genetic diversity in a threatened species

2.1 Introduction

It is estimated that 2000-3000 terrestrial vertebrate species will require captive breeding over the next 200 years to save them from extinction (Frankham et al. 2002; Soulé et al. 1986; Tudge 1995). Already, many species have been preserved in captivity following their extinction in the wild, for example Prezwalski’s horse (Equus prezwalskii), the black-footed ferret (Mustela nigripes) and the Arabian oryx (Oryx leucoryx) (Frankham et al. 2002). For many other species, captive breeding programs have been established as a form of insurance against extinction in the wild. Captive populations provide a source of individuals for reintroductions and should be able to establish a self-sustaining wild population with high reproductive fitness and ample genetic diversity (Frankham et al. 2002). The International Union for Conservation of Nature (IUCN) recommends that captive populations be founded before wild populations drop below 1000 individuals. The advantages of establishing a captive population at this stage include using wild individuals with low inbreeding levels, minimising the impact of removing individuals from the wild and allowing sufficient time to develop suitable husbandry techniques (IUCN 2006).

The maintenance of genetic diversity is an important goal in captive breeding for several reasons. First, genetic diversity supplies the raw material for adaptive change (Darwin 1859; Frankham et al. 2002). Maintaining the adaptive potential of any

Chapter 2 Genetic diversity in contrasting captive breeding programs 46 population, wild or captive, ensures the population has the greatest chances of coping with environmental change. Second, higher levels of heterozygosity have been associated with higher fitness in plants, vertebrates and invertebrates (Reed & Frankham 2003). Third, inbreeding increases homozygosity and can lead to an accumulation of deleterious alleles and reduced fitness i.e. inbreeding depression (Frankham et al. 2002). Often captive breeding programs are founded with a small number of individuals, which can result in a loss of genetic diversity, inbreeding and reduced fitness. In turn, this will reduce a population’s adaptive potential thus lowering the likelihood of a population’s long-term survival (Ralls & Ballou 1986; Reed & Frankham 2003). The amount of genetic diversity lost over time is determined not so much by the number of individuals present in a population (N, census size), but by the genetically effective population size (Ne) (Wright 1969). Therefore Ne is an important consideration for management strategies.

Several management strategies have been recommended to retain maximum levels of genetic diversity and minimise levels of inbreeding in captive populations, such as the ‘maximum avoidance of inbreeding’ (MAI) and ‘minimising kinship’ (MK) strategies.

MAI involves the equalisation of family sizes and the doubling of Ne (Frankham et al. 2002). MK involves managing pedigreed populations so that the matings are made between individuals that are the most distantly related in the population. The latter method has been shown to be more effective in maintaining genetic diversity than alternative techniques such as random choice of parents or MAI, and also helps reduce unequal representation of genotypes from founder individuals (Montgomery et al. 1997). Despite this, no significant difference in reproductive fitness has been detected between MAI and MK techniques when experimentally tested using 40 replicate populations of the vinegar fly (Drosophila melanogaster) managed over four generations (Montgomery et al. 1997). This study compares the genetic diversity of two captive breeding programs of the greater bilby (Macrotis lagotis) that have implemented different management strategies, unmanipulated mating and minimising kinship.

Introducing novel genotypes into a population (‘genetic rescue’) through translocation and exchange of individuals between different captive breeding programs can also help

Chapter 2 Genetic diversity in contrasting captive breeding programs 47 alleviate the effects of inbreeding and loss of heterozygosity (Bryant et al. 1999; Spielman & Frankham 1992). ‘Genetic rescue’ is based on the assertion that immigrants into a population can introduce new genetic variation and increase the fitness in populations that are experiencing inbreeding depression (Darwin 1883; Ingvarsson & Whitlock 2000; Tallmon et al. 2004; Whitlock et al. 2000; Wright 1931, 1940). This has been tested in recent studies in both natural and experimental populations (Tallmon et al. 2004). For example, a reduction in the proportion of stillborns, increase in genetic diversity, recruitment and population growth in a population of adders (Vipera berus) has been attributed to the introduction of immigrants (Madsen et al. 1999). The improved fitness is thought to be largely due to heterosis in the offspring that arises from matings between immigrants and local individuals. However, if the immigrant is highly genetically divergent from the local population, this can lead to outbreeding depression and can result in a subsequent reduction in fitness (Tallmon et al. 2004).

Over the past 200 years, more mammals have become extinct in Australia than any other country (Short & Smith 1994). Within Australia, captive programs for many species have been established with the aim of reintroduction into the wild, for example the western barred bandicoot (Perameles bougainville), Gilbert’s potoroo (Potorous gilbertii), banded hare-wallaby (Lagostrophus fasciatus), mallee fowl (Leipoa ocellata), greater stick nest rat (Leporillus conditor) and the greater bilby (Mawson 2004a). The greater bilby is Australia’s only remaining species of desert bandicoot and an iconic species for conservation. They are burrowing, nocturnal marsupials, belonging to the subfamily Thylacomyinae (Johnson 2002). During the past century bilby populations have been in rapid decline. Prior to European settlement, bilbies were widespread throughout the arid and semi-arid zones of Australia, but now are confined to just 20% of their former range (Southgate 1990). It has been suggested that competition with introduced herbivores (cattle, sheep and rabbits) and predation by introduced red foxes (Vulpes vulpes) and feral cats (Felis catus) have reduced populations as well as contributed to the failure of some reintroduction attempts (Moritz et al. 1997). The greater bilby is listed as ‘Vulnerable’ to extinction by the IUCN (IUCN 2006).

Chapter 2 Genetic diversity in contrasting captive breeding programs 48

As part of the national recovery plan for the greater bilby, several captive breeding programs have been established with the management goal of retaining at least 90% of current genetic variation for 100 years (Pavey 2006). Two such programs have been established in Western Australia (Figure 2-1): Return to Dryandra (hereafter referred to as Dryandra) and Peron Captive Breeding Centre (hereafter referred to as Peron). Both colonies were established within 12 months of one another, each founded with seven individuals, and the same sex ratio (3 females: 4 males). Dryandra is located in the Dryandra Woodlands, an area where bilbies have become locally extinct. Management has simulated the natural environment and mating was unmanipulated under semi-free ranging conditions. In contrast, Peron maintained a pedigree with the aim of minimising the mean kinship between individuals to ensure the maximum retention of genetic diversity. Both programs were developed by Western Australia’s Department of Environment and Conservation (DEC) and form a component of the largest wildlife recovery program ever undertaken in Australia, Western Shield, which aims to expand predator control and then reintroduce native animals to their former habitats (CALM 1999).

Francis Peron National Park

Dryandra

Chapter 2 Genetic diversity in contrasting captive breeding programs 49

To date, only two studies have examined the genetic variation of current and historically subdivided bilby populations (Moritz et al. 1997; Southgate & Adams 1993). There is no knowledge of how founder numbers and management strategies impact genetic diversity in captive breeding programs over time in a threatened marsupial. Such information is important for decisions influencing conservation and future management. The Dryandra and Peron captive colonies provide a unique opportunity to determine the impact of different management strategies on genetic diversity in this species, as well as being immediately applicable to captive breeding of any species in the world.

The aims of this study were to: (1) measure and compare genetic diversity within and between each captive colony overall, and compare these to wild populations; (2) monitor temporal changes in genetic diversity; and (3) assess the reliability of studbook estimates of genetic diversity and inbreeding compared to those calculated from microsatellite data.

2.2 Materials and Methods

2.2.1 Study populations

The Dryandra (32° 46’S, 116° 58’E) population was established in 1998. The breeding enclosure consists of 20 hectares (ha) of natural vegetation (Plate 2-1) surrounded by a 2.5m electrified fence (Plate 2-2). This is divided into two 10ha sections by a conventional fence, and one of these enclosures provides the basis of this study. The founder population consisted of five individuals of diverse origins from Kanyana (WA), and two wild caught individuals from north-western WA. Kanyana is a wildlife rehabilitation centre for wild bilbies that has a small breeding colony of individuals with a known pedigree. The founder sex ratio at Dryandra was three females to four males. Animals were semi-free ranging and mating was unmanipulated within their enclosure. The Peron population was established in 1997, and is located on 1050km2 of the Peron Peninsula (25° 42’S, 113° 32’E), which is severed from the mainland at its isthmus by a 3.4km electric fence. Animals were housed in small mesh-covered pens to accommodate breeding pens and large outdoor pens to accommodate small family groups. The founder population consisted of four individuals from Kanyana and three

Chapter 2 Genetic diversity in contrasting captive breeding programs 50 wild caught individuals, and all founders originated from north-western WA. The founder sex ratio was three females to four males and there was a known pedigree based on the breeding of the founder individuals. Animals were selectively bred to minimise kinship. Both populations had additional individuals translocated into the colony over time from Kanyana, or the wild, since establishment, as well as animals periodically being removed for reintroduction into the wild. In both populations, all individuals were identified using subcutaneous PIT tags (Destron, USA). Tissue samples were collected from individuals upon first capture as an independent juvenile or upon introduction into the colony and were stored in dimethyl sulfoxide (DMSO) prior to DNA extraction.

Plate 2-1 The Dryandra woodlands, Western Australia, that surrounds the Return to Dryandra captive breeding facility and serves as a reintroduction site for the locally extinct greater bilby (Macrotis lagotis).

Chapter 2 Genetic diversity in contrasting captive breeding programs 51

Plate 2-2 Predator proof fencing surrounding the Return to Dryandra captive breeding facility for threatened species in Western Australia.

2.2.2 Microsatellite genotyping

DNA was extracted from 2mm ear biopsies (Dryandra n = 216; Peron n = 266) using a standard salting-out method (Sunnucks & Hales 1996; Appendix 1). To enable a comparison between the captive populations and WA wild-caught individuals, frozen blood homogenates were obtained from the South Australian Museum from six wild caught bilbies from northern WA (lodged by R. Southgate). DNA was extracted from the blood homogenates using a FlexiGene DNA kit (Qiagen, Australia). Each individual was screened at nine polymorphic microsatellite loci (Bil02, Bil16, Bil17 Bil22, Bil41, Bil55, Bil56, Bil63 and Bil66) previously characterised from the greater

Chapter 2 Genetic diversity in contrasting captive breeding programs 52 bilby (Moritz et al. 1997). Genotyping was carried out using multiplexed PCRs, whereby fluorescently labeled primers enabled the simultaneous amplification of many targets of interest, using several pairs of primers in a single reaction. Two multiplex combinations were used (i) Bil55, Bil22, Bil16, Bil41, Bil17 and (ii) Bil02, Bil56, Bil63, Bil66. All loci were amplified in a “touchdown” PCR, whereby the initial annealing temperature was decreased at 1ºC increments and run for a total of ten cycles each. Ten micro-litre reactions were performed using a Multiplex PCR kit (Qiagen, Australia), 0.2μM of each primer and 60 – 80ng genomic DNA. Amplifications were carried out in a PTC-220 thermocycler (MJ Research, USA) using an initial HotStarTaq activation step at 95ºC for 15 min, a total of 50 cycles of 94ºC for 30 s, 60, 59, 58, 57, 56 and 55ºC for 1 min 30 s (10 cycles each), 72ºC for 1 min 30 s, and a final extension at 72ºC for 10 min. The PCR products were analysed in a 48 capillary AB 3730 DNA Analyser (Applied Biosystems, USA). The DNA fragments were sized and quantified using GeneMapper 3.7 (Applied Biosystems, USA).

2.2.3 Genetic analyses

The genotype file was checked for duplicate entries, scoring errors due to null alleles, short allele dominance and stutter bands using the program MICRO-CHECKER (van Oosterhout et al. 2004). Locus independence and Hardy-Weinberg equilibrium tests were conducted using GenePop 3.4 (Raymond & Rousset 2003) using a Markov chain method (1000 iterations). The statistical significance levels were corrected for multiple comparisons using sequential Bonferroni adjustments (Rice 1989). Genetic diversity was estimated for each population by calculating allelic diversity (AD), observed (Ho) and expected (He) heterozygosities using GENALEX 6.0 (Peakall & Smouse 2006). A Wilcoxon signed rank test was used to detect significant differences in diversity between Dryandra and Peron. To assess temporal changes in genetic diversity within each population, the dataset for each colony was further divided into annual ‘sub- populations’. Each sub-population represents every independent individual present in the population in a 12 month period. Due to the small number of individuals present in the first two years for both populations (1998 and 1999 for Dryandra (98/99) and 1997 and 1998 for Peron (97/98)), these years were pooled and examined as a single ‘sub- population’. A One-Way ANOVA using Tukey’s Post-Hoc Multiple Comparisons was used to test for differences in AD, Ho and He between years within each colony.

Chapter 2 Genetic diversity in contrasting captive breeding programs 53

The F-statistics, inbreeding coefficients (FIS) and genetic differentiation (FST), were calculated using Weir & Cockerham (1984) estimators from the microsatellite data using FSTAT 2.9.3.2 (Goudet 2001). The significance levels of FIS and FST were determined after 10 000 permutations. Population bottlenecks were tested for using BOTTLENECK 1.2.02 (Cornuet & Luikart 1996), assuming a two-phase model (TPM) with a 95% single-step mutations and 5% multiple step mutations, and a variance among multiple steps of 12. A Wilcoxon test was used as suggested for relatively low number of loci (Piry et al. 1999).

Microsatellite variation (AD and He) of both captive populations was compared to published data for the same microsatellite loci for wild M. lagotis populations from the Northern Territory (NT; n = 19) and Queensland (QLD; n = 27) (Moritz et al. 1997) using a Wilcoxon signed rank test in SPSS 15.0. To enable this comparison, Bil66, which was out of Hardy-Weinberg equilibrium for this study, was included in the analysis. Note Bil63 was also out of Hardy-Weinberg equilibrium for the wild QLD population (Moritz et al. 1997).

Population data for Peron was collected by DEC (WA) and maintained in SPARKS (ISIS 1992). Two pedigree-based statistics were generated using PM2000 (Pollak et al. 2002): (i) mean inbreeding coefficients (F) were calculated using SPARKS (ISIS 1992) and corroborated using PM2000; and (ii) retained gene diversity, that is, the proportion of heterozygotes expected in the descendant population under Hardy- Weinberg equilibrium assumptions relative to the founding population. Retained gene diversity is calculated by allocating two hypothetical alleles to each founder and running a gene-drop simulation analysis (1000 iterations) to determine what proportion of the founding genetic diversity is present in the current population according to Mendelian inheritance (Lacy & Ballou 2002). The studbook estimates were compared to the genotypic estimates of diversity using a Wilcoxon signed rank test. All statistical analyses were conducted in SPSS 15.0.

Chapter 2 Genetic diversity in contrasting captive breeding programs 54

2.3 Results

2.3.1 Genetic diversity

Eight of the nine loci amplified successfully (Bil17 failed to amplify). MICROCHECKER found five duplicate entries of individuals that had been sampled twice, and so the duplicates were removed from the dataset. All loci except Bil66 were in Hardy-Weinberg equilibrium. In both captive populations Bil66 showed a significant excess of homozygotes (p > 0.05) and evidence of null alleles was detected by MICROCHECKER. There was no evidence of allelic dropout in any locus. Bil17 and Bil66 were excluded from further analyses. The two captive bilby populations were polymorphic for all other loci. For both populations, across the seven loci, values of Ho varied between 0.394 and 0.823 and for He between 0.483 and 0.848 (Table 2-1). There were four to nine alleles per locus. One private allele was identified in the Dryandra population and nine were identified in the Peron population. The annual population allele frequency data for Dryandra and Peron are presented in Appendix 2 and 3, respectively. The overall allele frequency data comparing Dryandra and Peron are presented in Appendix 4.

Table 2-1 Summary of the overall genetic diversity calculated across all years from seven microsatellite loci for two captive bred populations of the greater bilby (Macrotis lagotis), Return to Dryandra (Dryandra) and Peron Captive Breeding Centre (Peron), Western Australia (mean ± se).

Parameter Dryandra Peron n 216 266 AD 6.4 (± 0.6) 7.6 (± 0.6) Range AD 4 – 8 5 – 9

Mean Ho 0.681 (± 0.047) 0.664 (± 0.055)

Range Ho 0.463 – 0.823 0.394 – 0.763

Mean He 0.709 (± 0.041) 0.758 (± 0.030)

Range He 0.483 – 0.813 0.628 – 0.848

Sample size (n), allelic diversity (AD), observed heterozygosity (Ho) and expected heterozygosity (He).

Chapter 2 Genetic diversity in contrasting captive breeding programs 55

2.3.2 Temporal changes in genetic diversity

Changes in allelic diversity (AD) and heterozygosity (He) over time for each population are shown in Figures 2-2 (a) and (b), respectively. The Dryandra colony showed an initial decline in AD, followed by a slight increase then plateau. The Peron colony showed an initial increase in allelic diversity then remained fairly constant until 2004, when a slight reduction in AD is evident. Dryandra has significantly lower AD than Peron (WRS z = -2.226, p = 0.026). The level of He in both populations has remained relatively constant since establishment, with a gradual increase in He evident in the Peron colony, but no significant difference to Dryandra (WRS z = -1.823, p = 0.068). In both colonies there were no significant differences between years for AD

(Dryandra: F6, 42 = 0.169, p = 0.984; Peron: F9, 60 = 1.062, p = 0.403) or He (Dryandra:

F6, 42 = 0.171, p = 0.983; Peron: F9, 60 = 1.404, p = 0.207).

The levels of inbreeding (FIS) over time are shown in Figure 2-3. The Dryandra population became increasingly outbred three years after establishment (mean FIS = 0.080; range = 0.016 – 0.123). This trend continued until 2003 when the levels of inbreeding began to increase. A significant FIS was detected in 2005 (p = 0.001). The positive values for Peron indicate more homozygotes being present in the population

(mean FIS = 0.136; range = -0.035 – 0.075), and FIS was significant in 2003, 2004 and 2005 (p = 0.001 each year). The number of individuals translocated into each population each year is indicated on this graph.

Chapter 2 Genetic diversity in contrasting captive breeding programs 56

(a) 10.0

9.0 )

AD 8.0

7.0

6.0 Allelic diversity ( diversity Allelic 5.0

4.0 97/98 98/99 2000 2001 2002 2003 2004 2005 2006 Year

(b)

1.0

) 0.9 e H 0.8

0.7

0.6

Mean heterozygosity ( 0.5

0.4 97/98 98/99 2000 2001 2002 2003 2004 2005 2006 Year

Figure 2-2 Genetic diversity over time for the greater bilby (Macrotis lagotis), captive breeding programs, Return to Dryandra (Dryandra, ) and Peron Captive Breeding Centre (Peron, ), Western Australia. (a) Mean allelic diversity (AD) and (b) mean heterozygosity (He).

Chapter 2 Genetic diversity in contrasting captive breeding programs 57

0.30 12 3 2 2 2 ) IS F 0.10

97/98 98/99 2000 2001 2002 2003 2004 2005 2006 -0.10 Inbreeding( coefficient

15 16 8 9 13 3 -0.30 Year

The overall genetic differentiation between Dryandra and Peron was low, but significant (FST = 0.056). There was also significant genetic differentiation within each captive colony over time (Dryandra, Table 2-2(a); Peron Table 2-2(b)). Dryandra showed no evidence of a genetic bottleneck in any year, whereas Peron showed evidence of a genetic bottleneck only in 2006 (p = 0.019).

Chapter 2 Genetic diversity in contrasting captive breeding programs 58

(a) Pop 2000 2001 2002 2003 2004 2005 98/99 0.002 0.009 0.013 0.008 0.011 0.013 2000 0.015 0.026* 0.018* 0.021* 0.020* 2001 0.001 0.007 0.011* 0.022* 2002 -0.001 0.004 0.017* 2003 -0.004 0.003 2004 -0.002 Significant differentiation indicated by *

(b) Pop 1999 2000 2001 2002 2003 2004 2005 2006 97/98 -0.007 0.007 -0.009 0.015 0.054* 0.067* 0.075* 0.083* 1999 -0.006 -0.005 0.019* 0.048* 0.055* 0.066* 0.067* 2000 0.000 0.030* 0.055* 0.058* 0.073* 0.070* 2001 0.016* 0.047* 0.060* 0.070* 0.075* 2002 0.004 0.017* 0.027* 0.047* 2003 0.004 0.022* 0.052* 2004 0.004 0.023* 2005 0.003 Significant differentiation indicated by *

2.3.3 Genetic diversity in comparison to wild populations

Dryandra had significantly lower levels of He (WRS z = -2.240, p = 0.025) and AD (WRS z = -1.761, p = 0.033) than the wild NT population. There were no significant differences in genetic diversity between Peron and NT, or between both captive populations and the wild QLD population (Table 2-3). Due to the small sample size of wild WA individuals (n = 6), the data could not be analysed statistically. The wild WA bilbies shared 100% of their alleles with the captive individuals: 78.9% were shared with Dryandra individuals and 100% were shared with the Peron population. Three

Chapter 2 Genetic diversity in contrasting captive breeding programs 59 alleles were present in the wild WA and Peron individuals that were not present in the Dryandra population.

Table 2-3 Genetic diversity for wild Northern Territory and Queensland populations, in comparison to the captive Western Australian (WA) populations of greater bilbies (Macrotis lagotis).

Population n Mean AD Mean He Northern Territory* 19 8.9 0.813 Queensland* 27 7.7 0.654 Dryandra (WA) 218 6.4 0.709 Peron (WA) 266 7.6 0.758

Sample size (n), allelic diversity (AD) and expected heterozygosity (He). *published data from Moritz et al. (1997).

2.3.4 Genetic diversity in comparison to studbook estimates

Overall the studbook estimates of genetic diversity for Peron were significantly higher than those estimated from the genotypic data (WRS z = -2.192 p = 0.028), and show the reverse trend (Figure 2-4(a)). The studbook estimates suggest that diversity slowly decreased over time, whereas the genotypic data shows a gradual increase. With higher estimates of genetic diversity, the studbook estimates of inbreeding were significantly lower than those calculated from the genotypic data (WRS z = -2.666 p = 0.008; Figure 2–4(b)), but indicated a similar trend in inbreeding over time.

Chapter 2 Genetic diversity in contrasting captive breeding programs 60

(a)

0.9 ) e H 0.8

0.7

0.6

Heterozygosity ( Heterozygosity 0.5

0.4 1999 2000 2001 2002 2003 2004 2005 2006 2007 Year

(b)

0.3 )

IS 0.2 F 0.1

0 1999 2000 2001 2002 2003 2004 2005 2006 2007 -0.1

Inbreeding ( coefficient -0.2

-0.3 Year

Figure 2-4 Comparison of the studbook (dashed line) and genotypic (solid line) estimates of (a) genetic diversity, and (b) inbreeding coefficients between 1999 and 2007 for the Peron Captive Breeding Centre (Peron) colony, Western Australia.

Chapter 2 Genetic diversity in contrasting captive breeding programs 61

2.4 Discussion

Maximising and maintaining genetic diversity is an important goal for captive breeding programs to ensure long-term population sustainability. This study found that (i) genetic diversity was maintained over time in both populations despite the different management strategies implemented; (ii) the supplementation of new individuals into the colonies provided a new source of genetic variation and aided the maintenance of genetic diversity; (iii) Peron had similar levels of genetic diversity to wild NT bilbies, but Dryandra had significantly lower diversity, and (iv) the studbook estimates of genetic diversity were significantly higher, and inbreeding significantly lower than those calculated from the genotypic data.

2.4.1 Genetic diversity in the captive bilby populations

The levels of allelic diversity and heterozygosity remained relatively constant in both captive breeding colonies, although Dryandra had significantly lower allelic diversity than Peron. In captive populations, genetic variation arises from the contribution of founding individuals, the introduction of immigrants and mutation (Ballou 1984). In small, closed populations mutation rates are likely to be too low to generate variation (Ballou 1984; Hedrick 2000). In addition, allelic diversity is strongly linked to population size (Frankham 1996) and small closed populations are expected to lose diversity due to random genetic drift (Lacy 1989). Consequently, we would expect the erosion of genetic diversity over time due to genetic drift in these captive bilby populations. However, we observed that genetic diversity was maintained over time in both the Dryandra and Peron captive breeding colonies, though at different levels.

Both captive bilby colonies started with similar levels of allelic diversity and in the year following establishment received a similar number of new individuals, yet after 12 months Peron had significantly higher allelic diversity than Dryandra. This difference possibly arose because those individuals introduced into Peron contributed more novel diversity than those introduced into Dryandra. Founding populations with unrelated individuals is important to minimise the loss of genetic diversity and inbreeding at the foundation for long-term conservation (Frankham et al. 2002; Gautschi et al. 2003). The strategy of minimising kinship through the use of a pedigree

Chapter 2 Genetic diversity in contrasting captive breeding programs 62 at Peron may have also contributed to the higher levels of diversity retained through initially equalising the founder contribution, that is, limiting the reproduction and genetic representation of the founder individuals (Ballou & Lacy 1995; Loebel et al. 1992). Equalisation of founder contributions is thought to diminish genetic drift by enlarging effective population size, thus resulting in a higher retention of allelic diversity (Lacy 1989). However, it is difficult to determine which management strategy has been more effective at maintaining diversity since both populations continued to translocate individuals into the population over time.

The translocation of new individuals into the populations may have buffered the effects of inbreeding and genetic differentiation within each colony. Between 1999 and 2005, 57 new individuals were translocated into the Dryandra colony. There was no significant differentiation between the founding and established population in 2005. Dryandra showed some evidence of significant (but low level) inbreeding in 2005, but there was no evidence of a genetic bottleneck. In contrast, Peron translocated fewer individuals (n = 21) between 1999 and 2004, experienced higher levels of significant inbreeding (2003 – 2005) and showed evidence of the beginnings of a genetic bottleneck by 2006. In addition, by 2003, Peron had significantly differentiated from its founding population. Similarly, a study of the bridled nailtail wallaby (Onychogalea fraenata) found captive-bred individuals and their wild-born offspring were significantly differentiated from the wild remnant populations within four generations as a result of rapid genetic drift arising from a small number of founders (n = 7) and subsequent loss of allelic diversity (Sigg 2006). They also showed that when individuals were removed from the captive breeding program for reintroduction into the wild, allelic diversity, but not heterozygosity decreased (Sigg 2006). If individuals are being removed from a population genetic diversity will decline if new genetic stock is not added to the population (Frankham 1996; Nei 1987; Sigg 2006).

There are substantial benefits for small, partially inbred populations when even a single unrelated animal is translocated into the population, for example increases in reproductive fitness (Bryant et al. 1999; Spielman & Frankham 1992). The strategy of transferring individuals among captive breeding facilities has proven to be beneficial for other species, for example, the bearded vulture, Gypaetus barbatus (Gautschi et al.

Chapter 2 Genetic diversity in contrasting captive breeding programs 63

2003), and a single immigrant increased heterozygosity and led to a rapid spread of new alleles in a re-founded population of the Scandinavian wolf, Canis lupus (Ingvarsson 2003; Vilà et al. 2003). Exchanging individuals with other facilities is an effective strategy for maintaining genetic diversity, but populations should continue to be monitored for genetic changes. There is evidence from salmon that suggests that breeding programs designed to supplement wild fisheries can reduce the fitness of natural populations through a reduced effective population size, mutation accumulation, and genetic adaptation to captivity (Araki et al. 2007; Goodman 2005; Heath et al. 2003; Wang & Ryman 2001).

2.4.2 Comparison to wild populations

In comparison with wild bilby populations, Peron had similar levels of diversity, but Dryandra had significantly lower heterozygosity and allelic diversity than the wild NT population but not wild QLD bilbies. When comparing the wild WA bilby genotypes with the captive populations, the results showed that Dryandra had not captured as much wild genetic variation as it potentially could have. A strategy to improve the representation of wild bilby diversity in Dryandra would be to source more wild caught individuals and/or exchange more individuals with Peron. However, the levels of diversity in both captive populations were within the range of values observed in other marsupial taxa including other ‘Vulnerable’ taxa (Table 2-4). As most captive populations are small, and/or founded with small numbers, it is not surprising that evidence suggests captive populations generally have lower genetic diversity than wild populations (Jiang et al. 2005; Kubota et al. 2008; Neveua et al. 1998). In the context of other captive and wild species, the genetic diversity in these captive populations of the greater bilby is reasonable.

Chapter 2 Genetic diversity in contrasting captive breeding programs 64

Table 2-4 Comparison of the genetic diversity in the captive greater bilby (Macrotis lagotis) populations with other wild marsupial populations of varying conservation status (after Bowyer et al. (2002)).

IUCN category No. populations AD He Least concern 8 12.0 – 5.2 0.86 – 0.66 Near threatened 4 11.1 – 5.3 0.85 – 0.60 Vulnerable* 3 8.9 – 6.0 0.86 – 0.65 Endangered 2 11.6 – 6.0 0.83 – 0.72 Critically endangered 1 1.8 0.27

Allelic diversity (AD) and heterozygosity (He). * Dryandra mean He = 0.709, AD = 6.4;

Peron He = 0.758, AD = 7.6.

2.4.3 Comparison to studbook estimates

In the Peron population there was a significant difference between studbook and genotypic estimates of genetic diversity and inbreeding. The studbook calculations based on pedigree data significantly overestimated genetic diversity, and consequently significantly underestimated levels of inbreeding. The studbook estimates also showed a declining trend in genetic diversity over time, but the reverse trend was apparent in the genotypic data. The Peron management program aimed to retain at least 90% of the genetic diversity found in the wild or from the source population. The estimates of genetic diversity from the studbook indicated that the Peron population was on target initially, but as time progressed, the diversity in the population decreased by 18.8% compared to the target of losing no more than 10%. In contrast, the genotypic data showed a significantly lower level of diversity in the population, but with genetic diversity increasing over time. As expected, the reverse was true for inbreeding. The studbook calculation for inbreeding was consistently lower than that estimated from the microsatellite data.

A difficulty that faces most captive managed populations is that the relationship amongst the founders is often unknown, and they are assumed to be unrelated. This assumption of unrelatedness between founder individuals will affect studbook calculations (Nielsen et al. 2007). If the founders are related, the studbook calculations

Chapter 2 Genetic diversity in contrasting captive breeding programs 65 are likely to overestimate genetic diversity, and consequently underestimate inbreeding in the population, as shown in this study. There is conflicting evidence in the literature regarding the accuracy of studbook estimates compared to genotypic estimates. While some studies have found them to be inaccurate (Haig et al. 1994; Jones et al. 2002; Morin & Ryder 1991), others have found studbook and genotypic estimates to be congruent and therefore suitable as a tool for genetic management (Nielsen et al. 2007; Wisely et al. 2003). As well as incorrect assumptions regarding the interrelationship of founders, these discrepancies can arise from fundamental differences in how the measures are calculated, inaccurate or incomplete studbook data, genotyping data, or uninformative genetic loci. These results highlight the importance of ensuring that founding individuals are unrelated prior to commencement and validation of studbook estimates of diversity with genotypic data as they form the basis for the genetic management of many threatened populations.

2.4.4 Implications for management

Species loss and the threat of extinction is a worldwide problem. Many species require captive breeding, and utilise studbook estimates of diversity to gauge the ‘genetic health’ of the captive population. Molecular genetics is a tool that has facilitated the evaluation of the effectiveness of management decisions and their impact on genetic diversity. In small, closed populations, founder contributions and new immigrants into a population are important for providing a new source of genetic variation. However, where possible, founders should be tested for relatedness. This will not only facilitate maximisation of genetic diversity, but also satisfy an underlying assumption of studbook diversity calculations which may improve the accuracy. In the case of the greater bilby, recruiting additional unrelated founder individuals and increasing the gene flow between captive bilby populations may assist in maintaining genetic variation in both the captive and the wild reintroduced populations. This will reduce the effects of genetic drift, genetic adaptation to captivity, inbreeding and genetic differentiation among populations. An additional benefit of exchanging individuals among populations could be further increasing the effective population size, thus maintaining the evolutionary potential of the greater bilby.

Chapter 2 Genetic diversity in contrasting captive breeding programs 66

The translocation of individuals between populations, zoos and wildlife institutions can provide ‘genetic rescue’ and be an effective conservation tool to mitigate the effects of small founder numbers, unequal founder contributions, inbreeding, genetic differentiation and genetic drift, but wildlife managers need to be aware of the potential problems that can occur. In conjunction with genetic management, breeding programs should utilise biological data to maximise genetic diversity. For example, a species mating system can influence the genetic structure of a population. If the mating system is polygynous and there is a large reproductive skew with only a few males contributing to the gene pool, an effective strategy would be to manipulate mating patterns by using a specific number of sires (Oyama et al. 2007). In a promiscuous mating system, more males participate in breeding, lowering the variance in male reproductive success which in turn would increase the effective population size and slow rates of inbreeding. An increase in the number of sires should lead to lower levels of relatedness within the population. Knowledge of the bilby mating system will help to further elucidate the relative influence of captive breeding strategies, genetic rescue and species mating systems on the maintenance of genetic diversity within captive breeding programs.

2.4.5 Conclusions

In summary, this study found the levels of genetic diversity were maintained over time in both captive breeding programs for the threatened greater bilby. Since both colonies translocated new individuals into the populations regularly, it was difficult to determine which strategy, if any, was more effective in maintaining genetic diversity. The introduction of new individuals helped mitigate the risk of genetic erosion, inbreeding and genetic differentiation that is expected to occur in small, closed populations founded with a small number of individuals. Although Dryandra had lower levels of diversity than the wild NT bilby population, the overall level of genetic variability in the captive bilby populations were comparable to that of wild bilby populations and other marsupial taxa. The studbook estimates of genetic diversity were overestimated in comparison to those calculated by the microsatellite data, and thus, levels of inbreeding were underestimated. The use of studbook genetic estimates should not be solely relied upon to evaluate the genetic health of the captive bilby population. Captive breeding programs should validate their studbook estimates with

Chapter 2 Genetic diversity in contrasting captive breeding programs 67 population specific genotypic data. This study highlights the importance of replenishing captive populations with new stock, especially post-animal removal for reintroductions.

Chapter 3 Genetic mating system, reproductive success and selection 68

Chapter 3 The genetic mating system, male reproductive success and selection on male traits in the Greater Bilby (Macrotis lagotis)

3.1 Introduction

Captive breeding programs are likely to become an increasingly important component of conservation strategies for terrestrial vertebrates in the future. Several species have already been preserved in captivity following their extinction in the wild, for example the California condor (Gymnogyps californianus) and Père David’s deer (Elaphurus davidianus) and it has been estimated that between 2000 and 3000 terrestrial vertebrate species will require captive breeding to prevent their extinction in the next 200 years (Frankham et al. 2002; Soulé et al. 1986; Tudge 1995). Captive breeding programs also serve as a form of insurance against wild extinction, providing individuals for wild reintroductions. Ideally, individuals selected for reintroduction should be physically healthy with a known high reproductive output and abundant genetic variation (Frankham et al. 2002), but this information is often not available because of both practical and financial constraints. In the absence of this information, individuals may be selected for release based on their body size as there is evidence in several species that males with a larger body size often have enhanced reproductive success (Andersson 1994; Birkhead 2000). Fundamental to successfully meeting this criterion is an understanding of the target species biology, including knowledge of their behaviour, reproduction, social organisation and mating system.

Chapter 3 Genetic mating system, reproductive success and selection 69

Knowledge of the mating system and parentage is crucial for the effective management of captive breeding programs in order to determine effective population sizes, detect inbreeding, genetic differentiation, and genetic diversity, all of which affect population growth and survival (Ralls & Ballou 1986; Reed & Frankham 2003). Parentage cannot be reliably determined using behavioural data alone. Advances in genetics have revealed inconsistencies in the presumed relationship between social organisation, parentage and mating system (Ambs et al. 1999; Coltman et al. 1999a; Worthington Wilmer et al. 1999), revolutionising how mating systems are understood. For example, around 95% of avian mating systems are classified as socially monogamous (Schwagmeyer & Ketterson 1999), however, genetic data has shown that extra-pair fertilisations occur on average in 13% (range 0 – 76%) of species (Westneat & Sherman 1997), leading to a better understanding of the evolution of sexual selection, mate choice and parental care. This study uses genetic markers to examine the mating system, male reproductive success and strength of selection on male morphological traits in a cryptic, threatened species that is the subject of a captive breeding and reintroduction program.

The diversity in mammalian mating systems is a product of the reproductive strategies of individuals and their environments, for example how many individuals males or females mate with and whether pair bonds form (Clutton-Brock 1989; Emlen & Oring 1977; Reynolds 1996). The classic mating system classifications are monogamy, polygyny, polyandry and promiscuity, though the precise definition of each can differ even within the field of evolutionary biology (Andersson 1994). Typically, monogamy is defined as the long-standing bond and exclusive relationship between a male and a female. Polygyny occurs when a single male mates with several females, and polyandry is the opposite, whereby a single female mates with several males (Wittenberger 1981). In a promiscuous mating system both males and females mate with more than one partner and no long term relationship forms (Wittenberger 1979). Polygyny commonly occurs in species where males do not provide parental care (i.e. most mammals). Consequently, males are able to invest their energy into competing for resources and mates (Emlen & Oring 1977). The ability to monopolise resources is a key factor in the intensity of sexual selection. Polygyny is expected to occur when there is spatial clumping of females in such a way that enables males to defend

Chapter 3 Genetic mating system, reproductive success and selection 70 multiple females from being accessed by other males (Clutton-Brock 1989; Emlen & Oring 1977) and is characterised by greater variance in male reproductive success than in females (Hoogland & Foltz 1982). Polyandrous mating systems are characterised by females pairing with multiple males during a single breeding season, but true polyandry has only been documented in a few mammalian species (Wittenberger 1979). In contrast, promiscuity generally occurs when males are unable to successfully monopolise access to females. It often occurs in species where female groups are unstable, males provide no parental care, and male home ranges are distributed throughout the home ranges of several females (Clutton-Brock 1989). Males are often unable to defend territories and therefore cannot monopolise access to females (Ramsay & Stirling 1986).

According to sexual selection theory, males ought to compete for access to receptive females, the most limiting resource for male reproductive success (Trivers 1972). There is evidence in numerous mammalian taxa demonstrating that a larger body size is advantageous when competing for access to receptive females (Clinchy et al. 2004; Fisher & Cockburn 2005; Fisher & Lara 1999). The male-biased sexual size dimorphism that exists in many mammalian taxa is often attributed to competition between individuals for reproductive opportunities, encouraging the evolution of secondary sexual traits (Andersson 1994; Birkhead 2000). Female preference for particular traits could potentially increase the strength of selection on male traits. For example, if females prefer larger males because their size indicated quality, strength or fighting ability (Andersson 1994; Bercovitch et al. 2003; Coltman et al. 2001; Schulte- Hostedde & Millar 2002), then we would expect strong selection for body size. Variation in body size may also influence male reproductive success through size- assortative mating. In several mammals, a positive correlation between mother’s body weight and litter size has been detected, for example possums (Julien-Laferriere & Atramentowicz 1990), opossums (Hossler et al. 1994), rodents (Kaufman & Kaufman 1987; McClure 1981; Myers & Master 1983; Svendsen 1964) and squirrels (Neuhaus 2000; Risch et al. 2007). Given the higher fecundity of larger females, positive size- assortative mating theoretically should differentially increase the fitness of larger males. Under this scenario, large male body size is not important per se, but rather male body size relative to female body size.

Chapter 3 Genetic mating system, reproductive success and selection 71

The greater bilby (Macrotis lagotis) is an iconic species for conservation in Australia. They are the sole remaining species of desert bandicoot and are listed as ‘Vulnerable’ to extinction (IUCN 2006). Bilbies are now restricted to 20% of their former distribution in the arid and semi-arid zones of Australia (Southgate 1990) due to habitat loss, competition with introduced species and predation (Moritz et al. 1997). Bilbies are distinguished by their soft, silky hair, long rabbit-like ears and long furry black and white tail. They are an omnivorous, nocturnal marsupial that live solitary or in pairs. They dig burrows up to two metres deep, and individuals may have up to a dozen burrows they use within their home range (mean: males = 316 ± 128ha; females = 18 ± 4ha) (Moseby & O'Donnell 2003). Bilbies are sexually dimorphic with males being larger than females (males: 800 – 2500g; females: 600 – 1100g). They are capable of breeding all year round, producing one to three offspring per litter, and up to four litters per year under ideal conditions. Males do not provide parental care (Southgate et al. 2000).

The bilby mating system is not well documented due to their secretive nature and the difficulties associated with wild observations. They are thought to be polygynous or promiscuous (Lee & Cockburn 1985). Johnson & Johnson (1983) monitored the behaviour of three adult males, two adult females and two female young in captivity (18m x 12m pen). They observed a rigid dominance hierarchy, maintained with little destructive aggression. Dominant males maintained access to all the well used burrows in the enclosure and chased subordinates out of the burrow (Johnson & Johnson 1983). Moritz et al. (1997) used genetic data to perform a parentage exclusion analysis for a colony of wild Queensland bilbies (n = 17). The partial pedigree constructed was suggestive of bilbies being strongly polygynous as one male mated with three females to sire seven of the eight offspring (Moritz et al. 1997).

Several bilby captive breeding programs have been established across Australia with the aim of wild reintroduction. Return to Dryandra (RTD) is a program located within the Dryandra Woodland, Western Australia (WA), an area where bilbies have become locally extinct. This program was developed by Western Australia’s Department of Environment and Conservation (DEC) and forms a component of the Western Shield

Chapter 3 Genetic mating system, reproductive success and selection 72 program, which aims to expand predator control and then reintroduce native animals to their former habitats (CALM 1999). Within the RTD population, the male selection criteria for reintroduction back into the wild is based the assumption that large males are monopolising mating opportunities within the large breeding enclosure. The removal of reproductively dominant males from the captive colony should theoretically aid the maintenance of genetic diversity both within the colony and within the reintroduced population, by enabling other captive males to gain fertilisations, and by releasing a healthy individual with a high reproductive potential. Therefore it is vital to understand the breeding biology and population genetics of the greater bilby to determine if this is an appropriate strategy for effective management of captive breeding and reintroduction programs.

This study aims to (i) clarify the mating system of the greater bilby using microsatellite markers; (ii) examine the variance in male reproductive success and determine whether a small number of males monopolise paternity; (iii) examine whether morphological traits are associated with male reproductive success; and (iv) determine if there is strong sexual selection acting on male traits to enhance their reproductive success. Most mating system research to date has been conducted in eutherian mammals, and marsupials have received little attention. This study makes an important contribution to the future conservation and management of the greater bilby, and adds to the growing body of scientific literature about threatened species.

3.2 Materials and Methods

3.2.1 Study population, data and sample collection

The RTD breeding facility is located in south-west WA (32º 46’S, 116º 58’E) and consists of two 10ha enclosed areas of natural vegetation, surrounded by a 2.5m electrified fence. This study was based on samples collected from one of the 10ha enclosures between 1999 and 2005. Food was supplied to the animals in nine small and four large feed hoppers in the enclosure. Animals were trapped quarterly between 1999 and 2005 using small cage traps covered with a Hessian sack and baited in the afternoon with a mixture of peanut butter, oats and anchovies. The traps were checked at 2200 and the following morning at sunrise. All animals were individually identified

Chapter 3 Genetic mating system, reproductive success and selection 73 using PIT tags (Destron, USA) inserted under the skin. A small ear biopsy (2mm) was collected from all individuals when they were first trapped. Each individual was weighed, and head (Plate 3-1(a)) and pes (foot) length were measured using vernier callipers. Additional measurements for scrotum length and breadth were also collected from males (Plate 3-1(b)), and the reproductive status was checked for females (Plate 3-2). Once the animals were processed, each individual was returned to their point of capture and released.

(a) (b)

Plate 3-1 Collecting morphological measurements from a male greater bilby (Macrotis lagotis), (a) head length (mm) and (b) testis length (mm).

Chapter 3 Genetic mating system, reproductive success and selection 74

3.2.2 Microsatellite genotyping

DNA was successfully extracted from small tissue biopsies from 216 individuals using a salting out method (Sunnucks & Hales 1996; Appendix 1). Each individual was screened at nine polymorphic microsatellite loci (Bil02, Bil16, Bil17, Bil22, Bil41, Bil55, Bil56, Bil63 and Bil66) previously characterised from the greater bilby (Moritz et al. 1997). Genotyping was carried out using multiplex PCR, whereby fluorescently labeled primers enabled the simultaneous amplification of many targets of interest, using several pairs of primers in a single reaction. Two multiplex combinations were used; (i) Bil55, Bil22, Bil16, Bil41, Bil17; and (ii) Bil02, Bil56, Bil63, Bil66. All loci were best amplified in a “touchdown” PCR, whereby the initial annealing temperature was decreased at 1ºC increments and run for a total of ten cycles each. Ten micro-litre reactions were performed using a Multiplex PCR kit (Qiagen, Australia), 0.2μM of each primer and 60 – 80ng genomic DNA. Amplifications were carried out in a PTC- 220 thermocycler (MJ Research, USA) using an initial HotStarTaq activation step at 95ºC for 15 min, a total of 50 cycles of 94ºC for 30 s, 60, 59, 58, 57, 56 and 55ºC for 1 min 30 s (10 cycles each), 72ºC for 1 min 30 s, and final extension at 72ºC for 10 min. The PCR products were analysed in a 48 capillary AB 3730 DNA Analyser (Applied Biosystems, USA). The DNA fragments were sized and quantified using GeneMapper 3.7 (Applied Biosystems, USA).

3.2.3 Parentage analysis

Locus independence and Hardy-Weinberg equilibrium tests were conducted using GenePop 3.4 (Raymond & Rousset 2003) using a Markov chain method (1000 iterations). The statistical significance levels were corrected for multiple comparisons using sequential Bonferroni adjustments (Rice 1989). Using MICROCHECKER, the genotype file was checked for duplicate entries, scoring errors due to null alleles, and stutter bands (van Oosterhout et al. 2004). Null alleles can result in parent-offspring mismatches, therefore loci with a high frequency of null alleles were excluded from paternity analysis.

Paternity and maternity were examined using CERVUS 2.0 (Marshall et al. 1998), a program that assigns paternity using a likelihood based approach. CERVUS calculates

Chapter 3 Genetic mating system, reproductive success and selection 75 a likelihood ratio (LOD) score for each candidate parent and assigns paternity to the most likely individual at a given statistical confidence, taking into account scoring errors, missing data and the proportion of candidate parents sampled. The program conducts simulations based on these input parameters to determine the significance levels of the LOD scores and estimate the expected rate of the successful parentage assignment in the population. Paternity may be assigned at 95% confidence (strict) criterion or 80% confidence (relaxed) criterion (Marshall et al. 1998). Offspring that were assigned at both the 80 and 95% confidence level were included in subsequent analyses.

To be able to assign an individual as a putative parent rather than an offspring, knowledge of individual’s age was required for the parentage analysis. Age class was calculated for all individuals using head length and body weight (Southgate 2005), and date of birth was estimated based on their age class. All individuals that were born in captivity were considered offspring (n = 145). To narrow down the number of candidate parents per offspring, the dataset was divided into annual ‘populations’. Each population represents every independent individual present in the population in a 12 month period. The assigned parentages were cross referenced with the candidate parent’s age and location. Of the total number of individuals relevant to this study (n = 232), 93% were genotyped successfully (n = 216). The proportion of loci sampled was 0.982, and the typing error was estimated as 0.006. Ten thousand iterations were performed for this simulation. Paternity was assigned to individuals with the highest LOD scores at the 95% and 80% statistical confidence levels. The paternity results were examined to assess whether males were successful in siring offspring in multiple years. Maternity was examined for evidence of matings with the same or different males within and between years. New individuals (total n = 57) were translocated into the population between 1999 and 2004. To examine whether females preferred new or ‘novel’ males introduced into the population, the proportion of sires that were immigrants was calculated.

Chapter 3 Genetic mating system, reproductive success and selection 76

3.2.4 Male reproductive success and selection analysis

Data Reduction using a Principal Components Analysis (PCA) was applied to determine the most suitable morphological variables for further analysis. A One-Way ANOVA using Tukey’s Post-Hoc Multiple Comparisons was used to test the relative importance of the various traits for male reproductive success by dividing the males into two subsets; sires and non-sires. A linear regression was used to establish whether male body weight could predict the number of offspring sired. Additionally, we tested for positive assortative mating between males and females with respect to body size. The body weight of successful males was regressed on the body weight of females with whom maternity was assigned. All statistical analyses were carried out using SPSS 14.0.

To assess whether variance in male reproductive success was arising from variance in fitness components (i.e., natural selection) (Arnold & Wade 1984), a selection analysis was performed. The strength of linear and non-linear selection upon male traits was calculated using a multiple regression based method (Lande & Arnold 1983) that accounts for the effects of correlation among traits. Briefly, male reproductive success was standardised so that the mean was equal to one, and the four fitness components, body weight (g), head length (mm), pes length (mm) and testicular volume (mm3), were transformed into Z scores for each male (n = 147). Testicular volume was calculated using the formula for an oblate spheroid, V = (/6).B2.L, where V = volume, B = breadth of testis and L = length of testis (Williamson et al. 1990). A multiple linear regression was fitted to the standardised traits and measure of reproductive success to calculate the vector of linear selection gradients (). A quadratic regression model was then applied to estimate the matrix of nonlinear selection gradients (). As nonlinear selection is often underestimated, a canonical analysis was performed to find the major axis of the response surface, the M matrix (Blows & Brooks 2003; Phillips & Arnold 1989). The appropriate gradients have been doubled in this analysis as it has been suggested that quadratic regression coefficients

(ii) obtained from statistical packages (including SPSS 14.0) need to be doubled to obtain the estimated quadratic selection gradients (Lande & Arnold 1983; Stinchcombe et al. 2008). The strength of the nonlinear selection along each of the eigenvectors (mi) of is given by their eigenvalues (i) using PopTools 3.0.3 (Hood

Chapter 3 Genetic mating system, reproductive success and selection 77

2008). Linear selection gradients of the eigenvectors (mi) of are given by . The major axis of selection was defined as the eigenvectors (mi) that display a significant level of linear and/or nonlinear selection. Selection on the major axis of the response surface was visualised using a spline three-dimensional surface plot in STATISTICA 7.0.

3.3 Results

3.3.1 Parentage assignment

MICROCHECKER found three duplicate individuals that were removed from the dataset. Hardy-Weinberg equilibrium was rejected for Bil66, as there was a significant (p > 0.05) excess of homozygotes and evidence of null alleles. There was no evidence of allele dropout for all loci. One locus (Bil17) failed to amplify successfully. Both Bil66 and Bil17 were excluded from further analyses. All remaining loci were polymorphic with the total number of alleles ranging from three to eight (mean =

6.4 ± 0.6) and the mean heterozygosity (He) estimates varying from 0.327 to 0.810 (mean = 0.709 ± 0.041). These data indicate there is an adequate level of genetic diversity to determine paternity. Allele frequency data are presented in Appendix 2.

Paternity was confidently assigned to 55 individuals at 80% statistical confidence, and 31 (56%) of these at the 95% confidence level. Paternity was not monopolised by a single or small group of males. Females did not show a preference towards immigrant males as there was no discrepancy between the proportion of immigrant males and their representation as sires within the population (19.6%). In any given year, 59.2 ± 9.3% of males in the population did not sire any offspring (Figure 3-1). Of the reproductively successful males, the majority sired one offspring (69.8 ± 3.6%), and 28.3 ± 1.8% sired multiple offspring (two or three) across one or two years, but not necessarily consecutively. One male sired six offspring in 2003. Maternity was confidently assigned to 55 individuals at 80% confidence and 25 (45%) of these at the 95% confidence level. Both paternity and maternity was confidently assigned to 20 offspring with 80% confidence. Only three of the 20 females had paternity assigned for multiple offspring (two each). Females did not repeatedly mate with the male that

Chapter 3 Genetic mating system, reproductive success and selection 78 sired their previous offspring i.e. we did not find any indication for mate fidelity. However, this needs to be interpreted with caution due to the small sample size (n = 6).

3.3.2 Morphological traits and male reproductive success

The PCA indicated that body weight, head length, scrotum length and width accounted for 82% of the variation in male reproductive success and were included in further analysis. Pes length was excluded as it contributed little to the model. Contrary to our predictions, sires and non-sires could not be distinguished from one another based on their morphological traits. There was no significant difference between sires and non- sires in body weight (F4, 143 = 1.135, p = 0.342), head length (F4, 143 = 0.590, p =

0.670), scrotum width (F4, 143 = 0.680, p = 0.607) or scrotum length (F4, 143 = 0.663, p = 0.619). The regression analyses further supported these results as there was no relationship between male body weight and the number of offspring sired in any given 2 year (r = 0.066; F1, 18 = 0.079; p = 0.782; Figure 3-2). In addition to the lack of evidence that morphological traits influence male reproductive success, there was no evidence of non-random (assortative) mating between males and females with respect to body weight. The regression revealed the body weight of females was not significantly correlated with that of the males who fathered their offspring (r2 = 0.003;

F1, 18 = 0.050, p = 0.826; Figure 3-3), suggesting random mating likely occurs in bilbies.

Chapter 3 Genetic mating system, reproductive success and selection 79

100

Percentage (%) of males 20

0 0123456 No. offspring sired

Figure 3-1 Percent (%) males siring offspring (± se) between 2000 and 2004 in a semi free-ranging captive greater bilby (Macrotis lagotis) population.

No. offspring sired sired offspring No. 2

0 0 500 1000 1500 2000 2500 3000 Body weight (g)

Figure 3-2 Relationship between male body weight (g) and the number of offspring sired between 2000 and 2004 in a semi free-ranging captive greater bilby (Macrotis lagotis) population.

Chapter 3 Genetic mating system, reproductive success and selection 80

1500 )

1000 Female weight (g weight Female

500 500 1000 1500 2000 Male weight (g)

Figure 3-3 Relationship between body weight (g) of male greater bilbies (Macrotis lagotis) and the females with which they sired offspring in a semi free-ranging captive population.

3.3.3 Selection analysis

To test whether selection was acting on male traits (summarised in Table 3-1) to enhance reproductive success, male reproductive success was standardised as a measure of fitness in a selection analysis. There was no evidence for overall strong linear or nonlinear selection acting on the male morphological traits measured in this study. This result was qualitatively the same, regardless of whether canonical rotation was applied to the data (Tables 3-2 and 3-3). However, there was no significant trend for non-linear selection on some aspects of the data. The canonical rotation of the matrix by symmetric Eigenanalysis of returned four eigenvectors (mi) representing the major axis of the response surface. The M matrix and their associated eigenvectors

(Table 3-3) show eigenvectors m1 and m4 to be under significant nonlinear selection.

Body weight and pes length had the heaviest loading for m1, and pes length had for m4.

The major axis of nonlinear selection, m1 and m4, indicates that there does not appear to be a single optimum phenotype associated with fitness (reproductive success) in

Chapter 3 Genetic mating system, reproductive success and selection 81 male bilbies. Males of various weights, with varying pes length had high fitness (Figure 3-4). Nevertheless, heavier males with a medium pes had higher reproductive success in this study. Body weight and pes length are correlated traits, and although the selection analysis is designed to cope with correlational data, this may explain why the overall selection was not significant.

Table 3-1 Male morphological traits (mean ± se) measured in the greater bilby (Macrotis lagotis) to examine the strength of selection.

Trait n Mean se Range Body weight (g) 147 1213.96 33.47 499 – 2483 Head length (mm) 147 104.88 0.82 85 – 138 Pes length (mm) 147 97.17 2.68 79 – 111 Testicular volume (mm3) 147 30.05 0.10 3 – 17

Table 3-2 Standardised linear gradients () and matrix of quadratic and correlational selection gradients () in the greater bilby (Macrotis lagotis). The multiple regressions describing selection were not significant (linear F4, 143 = 0.034, p = 0.875; nonlinear F4,

143 = 1.044, p = 0.387).

Trait Weight Head Pes Testis Weight -0.389 ± 0.53 1.240 ± 0.82 Head 0.640 ± 0.60 -0.440 ± 1.52 -1.037 ± 0.86 Pes -1.069 ± 1.74 -6.059 ± 3.97 5.197 ± 4.28 0.240 ± 0.21 Testis 0.435 ± 0.39 -1.289 ± 0.66 1.240 ± 0.74 3.528 ± 2.98 0.311 ± 0.24

Chapter 3 Genetic mating system, reproductive success and selection 82

Table 3-3 The M matrix of eigenvalues from the canonical analysis of presented in

Table 3-2. Overall the linear (i) and nonlinear (i) selection gradients along each eigenvector were not significant (F8, 139 = 0.994, p = 0.443), although two axis (m1, m4) were significant (*p < 0.05). Gradients are represented as ± standard error (se). Traits that contributed the most to the respective major axis are underlined.

M Selection

Weight Head Pes Testis i i m1 0.534 -0.376 -0.665 -0.362 0.054 ± 10.9 9.969 ± 1.21* m2 0.713 0.487 0.022 0.505 0.080 ± 0.17 -0.160 ± 0.06 m3 -0.120 -0.610 -0.168 0.765 0.124 ± 0.43 -1.250 ± 0.30 m4 0.439 -0.499 0.728 -0.170 -0.670 ± 1.29 -7.804 ± 0.98*

m1 m4

Figure 3-4 Visualisation of the fitness surface on the two major axes of nonlinear selection, m1 and m4. Reproductive success of male greater bilbies (Macrotis lagotis) was standardised to a mean of one.

Chapter 3 Genetic mating system, reproductive success and selection 83

3.4 Discussion

This study has shown that paternity was not monopolised by a single or small number of males in a semi-free ranging population of the greater bilby. In any given year, approximately half the males present in the population did not sire any offspring. Of the sires, approximately 70% sired one offspring and 30% sired multiple (two or three) offspring. There was no evidence for size assortative mating between males and females, further supporting the notion that mating among bilbies is likely to be random within the population across years. Based on these data, we suggest that the mating system adopted by the greater bilby is promiscuous. Sires and non-sires could not be distinguished based on their morphological traits and the selection analysis further supported these findings. There was no evidence for strong linear or non-linear selection on male morphological traits in these populations of the greater bilby. Since larger body size does not assist males to monopolise paternity, the use of body size as a selection criteria for wild release to allow other males to add to the gene pool is unnecessary.

3.4.1 Mating system

Wittenberger (1979) defined overlap promiscuity as promiscuous mating that occurs “between solitary individuals with overlapping home ranges or during brief visits by one sex to the home range or territory of the other”. It has been suggested that other solitary bandicoot species, including the southern brown bandicoot (Isoodon obesulus), long-nosed bandicoot (Perameles nasuta) and eastern barred bandicoot (P. gunnii), have overlapping promiscuous mating systems (Lee & Cockburn 1985). There are several lines of evidence from our data, combined with existing knowledge, that suggest the greater bilby has an overlapping promiscuous mating system. First, offspring paternity was not dominated by a single or small group of males. Male reproductive success was widely distributed among males with up to half of the males in any one year siring offspring (on average 12 of 21 males each year), indicating that males could not monopolise access to females. The primary determinant of a mating system is the ability of an individual to monopolise access to resources (Emlen & Oring 1977). In promiscuous species, variance in male reproductive success is low compared to the high reproductive skew observed in a polygynous mating system whereby a single or few males monopolise access to females (Johnstone 2000). For

Chapter 3 Genetic mating system, reproductive success and selection 84 example, in the promiscuous wood rat (Neotoma macrotis) there was no evidence for a strong skew in reproductive success and variance in reproductive success was similar for males and females (Matocq 2004). In contrast, in the polygynous eastern grey kangaroo (Macropus giganteus) a single, dominant male can monopolise access to several females and sire up to 86% of all offspring in a mob of kangaroos (Chapter 5).

Second, the greater bilby is a solitary, nocturnal species and there is no evidence to suggest they form long-term pair bonds. A more likely scenario is that the males are roving, and mate randomly with multiple females, given their large home ranges and patterns of burrow usage. Females are solitary and widely distributed, and although females maintain discrete home ranges from other females, they do overlap with other males (Moseby & O'Donnell 2003). Male home ranges are on average 18 times larger than female home ranges (Moseby & O'Donnell 2003). There is no evidence of female spatial clumping in the wild, and given their large home range it would be difficult for males to defend multiple receptive females simultaneously. Promiscuous species are often solitary, and in mammals solitary habits are associated with nocturnal activities (Eisenberg 1966). Females rely on secretive behaviour and concealment for protection, and an effective way of maximising concealment is to remain solitary. Promiscuity is advantageous in such species as males contribute little (or no) parental care or vigilance, both of which reduces the effectiveness of concealment tactics (Wittenberger 1979).

Third, other studies confirm that bilbies are unlikely to be territorial which further suggests promiscuous behaviour. Moseby & O'Donnell (2003) did not observe males to have exclusive access to females in their study which encompassed an area of 14 km2, and found different males visiting females on sequential days. However exclusivity of home ranges may not be indicative of a dominance hierarchy per se, but could arise from resource partitioning. The strict dominance hierarchy observed in captivity in the study conducted by Johnson & Johnson (1983) may have been a product of their enclosure size (12m x 18m), as the individuals may become territorial to gain access to limited resources. Moritz et al. (1997) showed that one male mated with three females to sire seven of eight offspring. The difference in paternity results

Chapter 3 Genetic mating system, reproductive success and selection 85 of this study compared to the high reproductive skew observed by Moritz et al. (1997) may be a result of our larger sample size and multi-year sampling.

Finally, theory suggests that a promiscuous mating system should evolve when the opportunity of gaining male parental assistance is not present or when the costs of taking advantage of them are too great (Wittenberger 1979). Male bilbies do not provide parental care (Southgate et al. 2000) and females are able to obtain the resources they require without male assistance. In addition, the high proportion of males siring offspring in any given year may also be an underestimate given that paternity was assigned to young when they were first caught as independent juveniles. The juvenile cohort does not include both pre- and post-weaning pouch young mortality which might have been sired by an even wider range of males than observed here. Paternity could not be assigned with confidence to all young in all years.

3.4.2 Male reproductive success

Reproductive success was fairly randomly distributed among male bilbies across years, the mean number of paternities per male was low (0.574) and a large body size did not provide males with a reproductive advantage. In species in which males can monopolise access to several females, sexual size dimorphism is more pronounced (Jarman 1983; Nunn 1999), and males can weigh up to two to three times that of the female; for example the eastern grey kangaroo, western grey kangaroo (M. fuliginosus), red kangaroo (M. rufus), whiptail wallaby (M. parryi), antilopine wallaroo (M. antilopinus) and agile wallaby (M. agilis) (Strahan 2002). Consistent with most medium-sized Australian marsupials that are secretive and solitary (Strahan 2002), the bilby shows little sexual size dimorphism (mean weight: males = 1213g n = 160 and females 1071g n = 160). In addition, the magnitude of sexual size dimorphism observed in the bilby is similar to that in other promiscuous species, for example meadow voles (Microtus pennsylvanicus) and snow voles (Chionomys nivalis) (Luque- Larena et al. 2004).

This study tested for positive size-assortative matings between male and female bilbies for two reasons: (i) the hypothesis that male body size would influence reproductive

Chapter 3 Genetic mating system, reproductive success and selection 86 success was tested by examining whether larger males were selecting females of a similar size; and (ii) based on the positive relationship found in several other species between a mother’s weight and litter size, we hypothesised a potential optimal strategy may be for larger males to mate with larger females to achieve a higher reproductive success. Paternity was random with respect to body size and males did not show a preference for mating with larger females. This is a similar result to that found in the promiscuous marsupial, the agile antechinus (Antechinus agilis) (Kraaijeveld-Smit et al. 2003).

3.4.3 Selection on male traits

The ability to monopolise resources is a key factor in the intensity of sexual selection. In light of the lack of relationship between male morphological traits and reproductive success it was not surprising that there was no evidence for strong selection acting on male morphological traits in the bilby. There was a weak trend for heavy males with medium pes to sire more offspring. These males may have a larger body size for their age, given their medium sized pes. But, given that there is no evidence for strong selection on these traits, mating success in this species may be more dependent on males being able to adopt a more conservative strategy that enables them to invest energy and resources into their search for mates rather than into growth, thus acting as a constraint on selection for a larger body size.

3.4.4 Implications for conservation and management

The data collected in this study enhances our understanding of the greater bilby mating system and this information should now be incorporated into management plans for bilby captive breeding programs. There are several captive breeding programs already established across Australia to conserve the greater bilby as it is vulnerable to extinction. In the RTD colony, larger males have been selected for reintroductions into the wild based on the assumption that larger males have sired the most offspring. While this assumption holds true for many species, this study found no evidence of size-related male reproductive success in the five years of data collected from this semi-free ranging population of bilby. These findings do not support the current

Chapter 3 Genetic mating system, reproductive success and selection 87 management practice of selecting larger males for reintroductions. Instead, selection of animals for reintroduction could be based on genetic analysis of reproductive success (if available) or, in the absence of these data, other parameters that are likely to reflect the chances of successful survival post-release, for example health and body condition, age. Alternatively, selection of animals for reintroduction could be random. In a promiscuous mating system more males participate in breeding, lowering the variance in male reproductive success which in turn, slows the rates of inbreeding, lowers the level of relatedness within the population and increases the effective population size (Frankham et al. 2002). Therefore the promiscuous mating system of the greater bilby is likely to aid the conservation of genetic diversity within captive breeding programs (Chapter 2).

There are still gaps in our knowledge and further research is required for a comprehensive understanding of the bilby mating system. Yet to be determined is whether females mate with multiple males and whether multiple paternities occur within each litter. This is a common feature in promiscuous species that produce multiple offspring per litter, for example the agile antechinus, A. agilis (Kraaijeveld- Smit et al. 2002), brown antechinus (A. stuartii) (Holleley et al. 2006), snowshoe hares (Lepus americanus) (Burton 2002) and the yellow-pine chipmunk (Tamias amoenus) (Schulte-Hostedde et al. 2004). Such information will improve our understanding of the role of sperm competition and female cryptic choice, and will enable us to assess alternative male tactics, all of which influence selection.

3.4.5 Conclusions

This paper presents the first comprehensive study of male reproductive success and morphological traits to clarify the mating system of the greater bilby. These data suggest the bilby mating system is promiscuous. Reproductive success was widely distributed across males consistently over a five years period. There was no evidence of size-assortative mating between males and females confirming that male body size is not important for gaining access to females, and mating among individuals is random. Consistent with these findings, there was no clear evidence for strong selection on male morphological traits, consistent with the other findings in this study.

Chapter 3 Genetic mating system, reproductive success and selection 88

The selection criteria for males to be removed from the captive colony for reintroduction into the wild does not need to be size based, but could be random or based on other traits related to post-release survival. It has to be acknowledged that the limitation of this study is that it was conducted using temporal data from a semi-free ranging captive population and caution must be applied when extrapolating these data to wild bilbies. However, based on the cryptic nature of the bilby and the difficulties associated with capturing them in the wild, this study makes an important contribution to the conservation and management of the greater bilby and other threatened species.

Chapter 4 Island population genetics 89

Chapter 4

Swimming wallabies? Genetics of three island populations of tammar wallabies (Macropus eugenii) in the Houtman Abrolhos Archipelago, Western Australia

4.1 Introduction

Increasing habitat fragmentation and population declines pose a major threat to the preservation of global biodiversity (Young et al. 1996). Many studies have shown island populations to have lower genetic variation than their mainland counterparts, and that island populations have a greater risk of extinction (Frankham 1997, 1998). Island populations are typically more susceptible than mainland populations to the effects of demographic and environmental stochasticity, and significantly more susceptible to the loss of genetic diversity due to random genetic drift and genetic bottlenecks as well as inbreeding depression (Eldridge et al. 1999; Frankham 1997, 1998). Small, isolated mainland populations face similar risks to those observed in island populations (Frankham et al. 2002). Inbreeding and loss of genetic diversity are of particular conservation concern because they not only reduce reproductive fitness, but increase the risk of extinction, and reduce a species ability to evolve and adapt to cope with environmental change (Frankham et al. 2002; Seymour et al. 2001). Records of extinctions since 1600 have shown that despite island species representing the minority of species in most groups, island species account for the majority of

Chapter 4 Island population genetics 90 extinctions (Frankham et al. 2002). Australia has the highest rate of recent mammal extinctions in the world (Short & Smith 1994), with many other species threatened and being actively managed for conservation. Many threatened species are either now confined to islands, or island populations constitute a significant population for the species.

Sea level rises during the past 6 000 – 15 000 years have resulted in the formation of many continental islands around Australia (Main 1961). This has resulted in the protection of many species from threatening processes occurring on the mainland, such as habitat loss and fragmentation, disease epidemics, competition and predation from introduced species, and stochastic environmental disturbances (Eldridge et al. 2004). In addition, islands are increasingly being utilised, both in Australia and internationally, as wildlife refuges for threatened mainland species (Daltry et al. 2001; Eldridge et al. 2004). For example Gilbert’s potoroo (Potorous gilbertii), one of the world’s most endangered mammals has been introduced to Bald Island, Western Australia in efforts to preserve the species (Courtenay & Friend 2003); northern quolls (Dasyurus hallucatus) have been introduced to Pobassoo and Astell Islands in the Northern Territory, Australia, to conserve the species as mainland populations are threatened by the spread of the introduced cane toad (Chaunus (Bufo) marinus) whose poison is deadly to mammals who predate them (Rankmore et al. 2008). South Island robins (Petroica australis australis) have been translocated to several offshore islands in New Zealand to establish populations protected from mammalian predators (Boessenkool et al. 2007) and the Southern sea otter (Enhydra lutris nereis) has been translocated to San Nicolas Island, California, to protect the vulnerable species from catastrophic events such as oil spills (Rathbun et al. 2000). However, the vulnerability of island populations to extinction may limit the suitability of islands as wildlife refuges. Studying island populations will make an important contribution to managing fragmented mainland populations as there are many parallels between the two, such as small population size and restricted gene flow.

The incorporation of genetics as a tool in wildlife management has provided some valuable insights, in particular the identification of Evolutionary Significant Units (ESUs) and Management Units (MUs) within ESUs (Hedrick et al. 2001c). An ESU is

Chapter 4 Island population genetics 91 a group of individuals with a similar genetic composition that demonstrate deep evolutionary divergence from other groups within the same species (Frankham et al. 2002). In contrast, MUs focus on identifying interactions between closely related populations within a species (Moritz 1994; Moritz 1999). The original definitions of ESUs incorporated two main elements: reproductive isolation (i.e. genetic distinctness) and ecological distinctness (Ryder 1986; Waples 1991). More recently, Moritz (1994) shifted the emphasis to genetic distinctness, suggesting ESUs should be recognised as ‘reciprocally monophyletic for mtDNA alleles and show significant divergence of allele frequencies at nuclear loci’ (Moritz 1994). Populations that have diverged in allele frequency but do not yet show reciprocal monophyly for mtDNA are also significant for conservation, and are referred to as MUs. Such cases represent populations that are functionally independent as they are connected by such low levels of gene flow. However, failure to understand a population’s history accurately could lead to an incorrect allocation of MUs (Sherwin et al. 2000).

The revolution in molecular biology (Sunnucks 2000) has led to an examination of the distribution of genetic variation within and among populations, as well as improving our understanding of population evolutionary history. This includes the use of mitochondrial, chloroplast and nuclear DNA. Mitochondrial DNA (mtDNA) has been widely used to address questions associated with population history, population genetic structure and patterns of gene flow. Recent reviews have raised concerns over studies that are based solely on mtDNA (Bazin et al. 2006; Edwards et al. 2005). Zink & Barrowclough (2008) sought to clarify this by reviewing the merits of mtDNA and nuclear phylogeographic studies using birds. They found mtDNA to be reliable for population inferences and that mtDNA and nuclear phylogeographical results should mostly concur, except when populations have been isolated for a shorter time, as nuclear markers are a lagging indicator for population structure due to their longer coalescence time (Zink & Barrowclough 2008). This current study addresses questions regarding population history, population structure and gene flow using mtDNA, autosomal and Y-linked microsatellite markers in an Australian marsupial, the tammar wallaby (Macropus eugenii).

Chapter 4 Island population genetics 92

The tammar wallaby is a medium-sized (5 – 10kg), herbivorous macropodid that lives in dense vegetation (Smith & Hinds 2002). Tammar wallabies are highly seasonal breeders, giving birth in late January – early February with a secondary peak late February – early March (Rudd 1994; Tyndale-Biscoe & Renfree 1987). They exhibit a post-partum oestrus approximately one hour after birth and the resultant conceptus grows to a 100-cell blastocyst before entering embryonic diapause. If the suckling young is lost during the period of decreasing day-length (between January and June in the Southern Hemisphere) the blastocyst will reactivate and birth will occur approximately 27 days later. If the young is lost after the winter solstice however, the blastocyst will remain in diapause until the following summer solstice (Tyndale-Biscoe & Renfree 1987). Tammars are endemic to coastal semi-arid regions of South Australia (SA) and Western Australia (WA; Figure 4-1). During the past 150 years, many populations have become greatly reduced or extinct due to habitat destruction, competition with exotic herbivores and predation by introduced species such as red foxes (Vulpes vulpes) and feral cats (Felis catus) (McKenzie & Cooper 1997; Smith & Hinds 2002). The tammar wallaby currently has a fragmented distribution (Figure 4-1) with a few disjunct populations surviving in mainland southwest WA and several offshore islands including the Abrolhos Archipelago, Garden Island (WA) and Kangaroo Island (SA). Two introduced populations thrive in New Zealand (Taylor & Cooper 1998).

The tammar wallaby poses interesting issues for wildlife management. They are listed as ‘Low Risk (near threatened)’ on a national scale (IUCN 2006), however there is a marked difference between the mainland and island populations. In SA they are extinct on the mainland and animals are being captive bred for reintroduction into their former range (DEH 2004). On mainland WA, the tammar wallaby was de-listed from the State Threatened Fauna List as a result of their recovery under the Western Shield conservation program which controls introduced predators (Kinnear et al. 2002; Possingham et al. 2003). In contrast, on several offshore islands they have become overabundant, for example on Kangaroo Island (SA) tammar wallabies are at such high densities, thousands are culled each year (Wright & Stott 1999).

Chapter 4 Island population genetics 93

Figure 4-1 Present and former distribution of the tammar wallaby (Macropus eugenii) (black = present distribution; grey = historic, (after McKenzie et al. (2006)) (A) indicates the location the Houtman Abrolhos Archipelago and (B) the Tutanning Nature Reserve, Western Australia.

The Houtman Abrolhos Archipelago (hereafter referred to as the Abrolhos) lies 60km offshore from Geraldton (WA) and became isolated from the mainland 11 500 years ago (Main 1961). The Abrolhos consists of at least 122 islands divided into three main island groups: Pelsaert, Easter and Wallabi Groups (Figure 4-2). The three largest islands of the Wallabi Group are the focus of this study, East Wallabi, West Wallabi and North Islands. West Wallabi Island (WWI) is the largest of the three (587 hectares (ha)) and became separated from the adjacent East Wallabi Island (EWI: 307ha) 6 000 years ago (Peltier 2004). EWI and WWI are geographically close (1.9km apart) to one another (GeoscienceAustralia 2005) and much of this distance is exposed at low tide as high tide is only an average of three to five metres deep (BOM 2008). The third and smallest island studied was North Island (176ha), which lies 14km northwest of EWI and WWI (Abbott & Burbridge 1995; GeoscienceAustralia 2005). WWI is of

Chapter 4 Island population genetics 94 particular significance as it is the location of the first recorded sighting of an Australian mammal by Europeans. The survivors of the infamous shipwreck The Batavia saw tammars in 1628, and the Dutch captain Francisco Pelsart recorded them in his journal in 1629 (Drake-Brockman 1963). EWI and WWI remain uninhabited, but fishermen reside on NI between mid-February and early June each year.

EWI (Plate 4-1(a)) and WWI (Plate 4-1(b)) are continental islands and tammar wallabies became stranded on these islands following sea level rise following the end of the last glacial maximum (Alexander 1922; Storr 1965). In contrast, NI (Plate 4- 1(c)) is a more recently formed sand island and the resident tammar wallaby population must have either naturally dispersed from EWI and/or WWI, or been introduced. The first recorded observation of a tammar wallaby on NI was in 1928 following their introduction (source population unknown) as a potential food source for stranded fisherman (Alexander 1922; Poole et al. 1991; Storr 1960). This introduction was thought to be unsuccessful and the animals subsequently were recorded as extinct (Storr 1960). In 1959, tammar mandibles were discovered on NI but it was not known whether they derived from the previous 1928 introduction or represent another introduction attempt. More recently, there have been at least two further attempts to establish tammar wallabies on NI (Poole et al. 1991) using animals sourced from EWI (C. Herbert, pers. comm.). The first was in 1983, when three animals were introduced. No animals were subsequently observed and the population was presumed extinct, so in 1985, five more animals were introduced (C. Herbert, pers. comm.). The sex ratio and reproductive status of the individuals introduced to NI is unknown. In a best case scenario, if the five individuals introduced in 1985 were all female, carrying pouch young and pregnant with a blastocyst, the founder population for NI could be a maximum 15 individuals (assuming 100% survival). The latest introduction appears to have been successful with a flourishing tammar population present on NI by 2002. The population has continued to increase aided by human- mediated modifications to the landscape on NI, which has included the increased availability of fresh water. Increasing concerns about overabundance and damage to the vegetation on NI resulted in a cull of ~ 800 animals in 2008 (A. Desmond, pers. comm.).

Chapter 4 Island population genetics 95

(a)

(b)

Chapter 4 Island population genetics 96

(c)

Plate 4-1 The harsh environment on (a) West Wallabi, (b) East Wallabi, and (c) North Islands in the Houtman Abrolhos Archipelago, Western Australia.

The aims of this study were to (i) assess the impact of long-term isolation and small population size on genetic diversity in each island population, and compare to the closest extant mainland population (Tutanning Nature Reserve), (ii) examine whether there is gene flow between EWI and WWI, (iii) confirm the local belief that animals on NI were introduced from EWI, and the effect of low founder numbers, and (iv) detect whether the animals in the Wallabi Group are of particular genetic or conservation significance. This is the first study to examine population genetics of tammar wallabies in the Houtman Abrolhos Archipelago, WA.

4.2 Materials and Methods

4.2.1 Study populations

This study focuses on the Wallabi Group (Figure 4-2), the most northerly island group within the Houtman Abrolhos Archipelago. Tammar wallabies are present on the three largest islands in the Group: East Wallabi Island (28º 26' S 113º 43' E), West Wallabi Island (28º 28' S 113º 41' E) and North Island (28º 18' S 113º 35' E).

Chapter 4 Island population genetics 97

4.2.2 Sample collection and DNA extraction

Trapping took place between 2006 and 2008 using a combination of wire cage (380 x 380 x 760mm) and mesh Thomas traps (N. Thomas, pers. comm.). Two trap-lines of between 10 and 25 traps, spaced ~50m apart, were set on each island and baited at dusk with a mixture of peanut butter, vegetable oil, oats and chopped apple. Traps were checked daily at dawn and the captured animals were removed and placed into Hessian bags. Traps were then left closed until evening. All animals were identified by a unique numeric code (subcutaneous microchip; Allflex®, NSW) and a small ear biopsy (2mm) was collected prior to the animal’s release. Ear biopsies were stored in 80% ethanol (EtOH) before the DNA was extracted using the standard salting out method (Sunnucks & Hales 1996; Appendix 1).

N North Island

East Wallabi Island

West Wallabi Island

1.0 km

Figure 4-2 Geographic relationship of East Wallabi, West Wallabi, and North Islands, within the Wallabi Group of the Houtman Abrolhos Archipelago, Western Australia (Hesperian 2007).

Chapter 4 Island population genetics 98

4.2.3 Microsatellite amplification and screening

All sampled individuals were genotyped using 10 polymorphic autosomal microsatellite loci characterised from the tammar wallaby (T19-1, T46-5, T31-1, T3- 1T, Me2 and Me14) and eastern grey kangaroo (M. giganteus) (G16-1, G26-4, G31-1, G20-2) (Taylor & Cooper 1998; Zenger & Cooper 2001; Zenger et al. 2003). Genotyping was carried out using fluorescently labeled primers in multiplexed ‘step- down’ PCRs. Two multiplex combinations were used (i) G31-1, T46-5, T31-1, G16-1, T3-1T and (ii) G26-4, G20-2, Me2, Me14, T19-1. Ten microlitre (μl) reactions were performed using a Multiplex PCR kit (Qiagen, Australia), 0.2μM of each primer and 30 – 40ng genomic DNA. Amplifications were carried out in a PTC-220 thermocycler (MJ Research, USA) using an initial HotStarTaq activation step at 95ºC for 15 min, a total of 50 cycles of 94ºC for 30 s, 64, 60, 56, 53 and 50ºC for 1 min 30 s (10 cycles each), 72ºC for 1 min 30 s, and final extension at 72ºC for 10 min. Males were also genotyped for an additional four Y-linked microsatellite loci, MeY01, MeY27, MeY28 and MeY37 (Macdonald et al. 2006), using 10μl reactions (as described above) and altered PCR conditions, whereby the temperature range used was a ‘touchdown’ cycle that decreased in 1°C increments between 64°C and 59°C. These paternally inherited markers were used to examine male-specific population histories. MeY37 amplified two loci and are referred herein to as MeY37A and MeY37B. All PCR products were analysed in a 48 capillary AB 3730 DNA Analyser (Applied Biosystems, USA). The resultant DNA fragments were sized and quantified using GeneMapper 3.7 (Applied Biosystems, USA).

4.2.4 Mitochondrial DNA amplification and screening

The mtDNA control region was amplified using marsupial-specific primers, L15899M and H16498M (Fumagalli et al. 1997). PCR reactions were performed in 20μl reactions containing 30 – 40ng genomic DNA, 12.2l milli-Q H2O, 2.0μl 10 x buffer,

2.0μl MgCl2 [25mM], 2.0μl dNTPs [2mM dCTP, dGTP, dTTP, lmM dATP], 0.4μl forward primer [25mM], 0.4μl reverse primer [25mM], and 0.04μl taq polymerase (Qiagen, Australia). Amplifications were carried out in a PTC-220 thermocycler (MJ Research, USA) with an initial 94ºC denaturation for 5 min, followed by 30 cycles of 94ºC for 30 s, 60ºC for 15 s and 72ºC for 30 s, and a final extension at 72ºC for 5 min.

Chapter 4 Island population genetics 99

The PCR products were then cleaned using the ExoSAP-IT (USB, USA) protocol. Briefly, 5μl of PCR product was transferred into sequencing plate, 1μl ExoSAP-IT was added per sample. The plate was covered with a mat, vortexed, spun down and placed in the PCR machine for 30 min (37°C for 15 min and 80°C for 15 min). This was followed by a Cycle Sequencing Big Dye Reaction (ABI PRISM®, USA) to enable fluorescence-based cycle sequencing reactions. A 14μl reaction mix consisting of forward primer [5mM], 5x buffer, milli-Q H2O and Big Dye terminator reagent was added to the Exo-SAPped samples. The samples were vortexed, spun down and placed in the PCR machine for 39 cycles of 96°C for 10 s, 50°C for 5 s and 60°C for 4 min. An EtOH/EDTA precipitation procedure was then performed. This involved the addition of 5μl [125mM] EDTA to each sample, 60μl of 100% EtOH was then added to each sample. The plate was sealed with a mat, vortexed and spun down. The plate was left at room temperature for at least 15 min in the dark. The products were centrifuged at 3000 rpm for 30 min at 4°C. The supernatant was discarded immediately and each sample was washed with 60μl of 70% EtOH. The plate was sealed with a mat, vortexed and centrifuged at 1650 rpm for 15 min at 4°C. The supernatant was discarded immediately by placing the plate upside-down on kimwipes and centrifuging at 190 rpm for 1 min at 4°C. The samples were placed in a dark fume hood to dry. The PCR products were analysed in a 48 capillary AB 3730 DNA Analyser (Applied Biosystems, USA). The mtDNA fragments were assembled and quantified using Sequencher 4.8 (Gene Codes Corporation, USA).

4.2.5 Estimates of autosomal and Y-linked microsatellite diversity

Locus independence and Hardy-Weinberg equilibrium tests were conducted using GenePop 3.4 (Raymond & Rousset 2003) via a Markov chain method (5000 iterations). The statistical significance levels were corrected for multiple comparisons using sequential Bonferroni adjustments (Rice 1989). Genetic diversity was estimated for each island population by calculating the allelic diversity (AD), observed (Ho) and expected (He) heterozygosities using GENALEX 6.0 (Peakall & Smouse 2006). The effective inbreeding coefficient (Fe, Wright’s fixation index) was calculated from the equation:

Fe = 1 – HIS / HM

Chapter 4 Island population genetics 100

Where HIS represents heterozygosity for island populations and HM represents heterozygosity for mainland populations (Frankham 1998), and a One-Sample t-test was used to test whether Fe differed significantly from zero. Estimates of genetic differentiation (FST) were calculated using FSTAT 2.9.3.2 (Goudet 2001) and significance was tested after 10 000 permutations using the Weir & Cockerham (1984) method. The differences in AD, Ho and He among populations were assessed using a Wilcoxon signed rank test. To test the hypothesis that island populations have lower genetic diversity than comparable mainland populations, a Wilcoxon signed rank test was conducted comparing the AD and He of each island population with the WA mainland population, Tutanning Nature Reserve (n = 10), typed for the same loci. All statistical analyses were conducted in SPSS 15.0. The Y-linked microsatellites were characterised by the alleles and haplotypes present in each population since there was insufficient haplotypic diversity to analyse statistically. The mainland WA population was not typed for the Y-linked microsatellites as the majority of the individuals sampled were female (n = 7).

BOTTLENECK 1.2.02 (Cornuet & Luikart 1996) was used to test whether a genetic bottleneck has occurred in any of the populations. A Wilcoxon test was used as suggested for relatively low number of loci, under the two-phase model (TPM) with a 95% single-step mutations and 5% multiple step mutations, and a variance among multiple steps of 12 (Piry et al. 1999) .

4.2.6 Estimates of mtDNA haplotypic diversity

ARLEQUIN 3.11 (Excoffier et al. 2005) was used to calculate haplotypic diversity (h), nucleotide diversity () and to assess population differentiation with the fixation index

ST, an estimator that includes information on haplotypic frequency and molecular distance using a pairwise distance method with a gamma distribution of 0.5. Molecular diversity indices theta-S (S) and theta-k (k) were also estimated in ARLEQUIN 3.11 using different aspects of the genetic data to estimate 2Nfe whereby Nfe represent the female effective population size and , the mutation rate. Theta-S is based on the relationship between the number of polymorphic sites and theta-k is based on the number of distinct lineages. Both measures are sensitive to the effects of lineage sorting during recent demographic history. An analysis of molecular variance

Chapter 4 Island population genetics 101

(AMOVA) was conducted in ARLEQUIN 3.11 to assess the population differentiation using haplotypes between two groups: EWI, and a combined WWI and NI since NI shared 100% of alleles with WWI. The estimates of the Abrolhos tammar wallaby population’s haplotypic diversity were not compared with mainland WA as that is the subject of a separate investigation (Eldridge et al. unpublished data).

4.2.7 Population structure and gene flow

Population subdivision based on mtDNA was also tested in ARLEQUIN 3.11 using three estimates of variance: among groups, among groups within populations, and within populations (Excoffier et al. 2005). To examine the presence of population structure, gene flow and assign a source population for NI, a Bayesian model-based clustering method was used within the program STRUCTURE 2.2 using autosomal microsatellite data (Pritchard et al. 2000). An admixture model was used with the alpha to be inferred from the data. Lambda, the parameter of the distribution of allelic frequency was set to one. The burn-in length was 100 000 iterations and the MCMC (Markov chain Monte Carlo) was set to 1 000 000. The range of possible populations (K) tested was from one to nine. The true number of populations (K) was identified using two methods: (i) K was identified using the maximal value of the estimated log probability of the data, Ln P(D), at each step of the MCMC, generated by STRUCTURE (Pritchard et al. 2000); (ii) K was estimated using an ad hoc statistic K based on the rate of change in the log probability of data between successive K values (Evanno et al. 2005). The latter method was implemented in addition to the traditional STRUCTURE method as a comparison as Evanno et al. (2005) found it to provide a more accurate estimation of the number of clusters, K.

4.2.8 Phylogenetic analysis

The mtDNA control region sequences (642 base pairs (bp)) were aligned using ClustalX (Thompson et al. 1997). For phylogenetic analysis, mainland WA and Kangaroo Island (SA) mtDNA sequences were also used (Zenger & Eldridge, unpublished data). These, along with the Abrolhos sequences, were assembled and trimmed to 595 bp using Sequencher 4.8 (Gene Codes Corporation). Phylogenetic trees of the mtDNA fragments were reconstructed using Neighbour-Joining (NJ), Maximum

Chapter 4 Island population genetics 102

Parsimony (MP) and Maximum Likelihood (ML) methods in PAUP 4.0 (Swofford 2002), and also by the Bayesian inference method implemented in MRBAYES 3.1.2 (Huelsenbeck & Ronquist 2001). Homologous sequences from an eastern grey kangaroo (Macropus giganteus) and western grey kangaroo (M. fuliginosus) (GenBank Accession numbers: AF443160 and AF443174, respectively), were used as an outgroup. To determine the most appropriate model for these data, 56 tests of evolutionary models were carried out in PAUP 4.0 and analysed using MODELTEST 3.7 (Posada & Crandall 1998). The recommended model was the HKY model (Hasegawa et al. 1985) with relative frequencies A (0.3943), C (0.2569), G (0.0662) and T (0.2826) and –lnL = 799.2745. Rate variation across sites was modeled using a gamma distribution and the proportion of sites was modeled as invariant. The NJ analysis used an uncorrected p, and gamma distribution = 0.5 for 10 000 replicates. The ML analysis used a branch-and-bound search for maximum parsimony. The model was used for the ML analysis in PAUP 4.0 using 1000 replicates. For the Bayesian analysis, the HKY model was also used in MRBAYES and run for 10 000 generations with four chains, sampling tress at every 100 generations, with a burn-in of 1000. A 50% majority consensus tree was obtained from the last 9000 trees. The robustness of each branch was evaluated by the bootstrapping method with 1000 replicates in all the Bayesian, MP and ML analyses. Heuristic searches were conducted for bootstrapping.

4.2.9 Computer simulations

The rate of loss of genetic diversity on each island was predicted using the simulation program GENELOSS (England & Osler 2001). Effective population size (Ne) for each population was estimated (50 000 iterations) from microsatellite data using ONeSAMP which uses summary statistics calculated from the data in an approximate Bayesian framework (Tallmon et al. 2008). In GENELOSS, a Monte Carlo sampling method (1000 iterations) was used to simulate the effects of population bottlenecks on both allelic diversity (AD) and heterozygosity (He). Simulations were conducted for EWI and WWI with Ne = 40, Ne = 60 and Ne = 100, all of which are greater than their current estimated Ne (see Results section 4.3.5). The genetic diversity of the Abrolhos tammar wallabies was assumed to start with similar levels of genetic diversity to the current mainland WA (Tutanning) population (n = 10; AD = 10.0, He = 0.822).

Chapter 4 Island population genetics 103

4.3 Results

In total, 101 individuals were trapped and genotyped (EWI n = 35, males = 20, females = 15; WWI n = 30, males = 16, females = 14; NI n = 36 males = 19, females = 17).

4.3.1 Autosomal and Y-linked microsatellite diversity

Nine of the 10 autosomal microsatellite loci amplified successfully. One locus (T19-1) could not be amplified consistently for all individuals and was excluded from further analysis. EWI and WWI populations were polymorphic for all autosomal microsatellite loci however NI was monomorphic for one locus (G20-2; Appendix 5). All three populations were in Hardy-Weinberg equilibrium at all loci (p > 0.05). Genetic diversity (AD, Ho, and He) was highest in the WWI population, and lowest on NI

(Table 4-1). There were no significant differences in Ho and He among the populations (p > 0.05). NI had significantly lower AD than both EWI (WRS z = -2.271, p = 0.023) and WWI (WRS z = -2.220, p = 0.026). Across the nine autosomal microsatellite loci, WWI had eight private alleles and EWI had 17 (Table 4-1). NI had no private alleles; all were shared with WWI. All three Abrolhos populations were significantly inbred.

Fe was significantly different from zero for all populations (p < 0.05). Although Fe, was highest in NI (Fe = 0.551), a similar level was detected in EWI (Fe = 0.545), with

WWI having the lowest (Fe = 0.377). All three populations were significantly differentiated from one another (FST range = 0.118 – 0.520; p = 0.008; Table 4-2), and the mainland population (p = 0.008). The Wilcoxon test in BOTTLENECK detected no evidence of a genetic bottleneck occurring in any population. The mainland WA population of tammar wallabies had significantly higher AD, Ho and He than all three Abrolhos populations (Table 4-1; AD: WRS z = -2.023 to -0.041, p = 0.041 – 0.043;

Ho: WRS z = -1.761 to – 0.826, p = 0.040 – 0.048; He: WRS z = -2.023, p = 0.043).

Chapter 4 Island population genetics 104

Table 4-1 Summary of autosomal microsatellite diversity indices for the East Wallabi Island (EWI), West Wallabi Island (WWI) and North Island (NI) tammar wallaby (Macropus eugenii) populations, and compared to a mainland Western Australian population.

EWI WWI NI Mainland n 35 30 36 10 AD (± se) 3.7 (± 0.6) 4.1 (± 0.7) 2.8 (± 0.5) 10 (± 1.6) Range AD 2 – 7 2 – 8 1 – 6 6 – 15 Proportion PA 0.68 0.32 0.00 –

Ho (± se) 0.382 (± 0.100) 0.491 (± 0.095) 0.386 (± 0.087) 0.709 (± 0.099)

Range Ho 0.029 – 0.800 0.100 – 0.800 0.000 – 0.722 0.444 – 0.900

He (± se) 0.374 (± 0.099) 0.512 (± 0.086) 0.369 (± 0.086) 0.822 (± 0.038)

Range He 0.028 – 0.782 0.156 – 0.809 0.000 – 0.765 0.690 – 0.910

Sample size (n), allelic diversity (AD), observed heterozygosity (Ho), expected heterozygosity (He), and proportion of private alleles (PA) within the Abrolhos populations.

EWI WWI NI Mainland EWI 0.000 0.501* 0.520* 0.366* WWI 0.000 0.118* 0.324* NI 0.000 0.344* Mainland 0.000

A total of 55 males were successfully genotyped for four Y-linked microsatellite loci. One locus, MeY27, failed to amplify consistently and was not included. Two loci were polymorphic (MeY01 and MeY28) and two were monomorphic (MeY37A and MeY37B; Table 4-3). A total of seven alleles and three haplotypes were identified

Chapter 4 Island population genetics 105 across all three populations (Table 4-3; Figure 4-3). EWI had one unique haplotype; WWI had two haplotypes, one of which was shared with NI (Figure 4-3). There was insufficient haplotypic diversity to conduct further analysis. Allele frequency data for the four Y-linked microsatellite loci are presented in Appendix 6.

Number of alleles Allele range (bp) Locus EWI WWI NI EWI WWI NI MeY01 1 1 1 310 314 314 MeY28 1 2 1 325 329 - 331 331 MeY37A 1 1 1 151 151 151 MeY37B 1 1 1 169 169 169 Total 4 5 4

2.0 1.8 1.6 1.4 1.2 1.0

Frequency 0.8 0.6 0.4 0.2 0.0 YA YB YC Y-linked haplotype

Figure 4-3 Distribution of Y-linked microsatellite haplotypes across the three Abrolhos Island tammar wallaby (Macropus eugenii) populations, East Wallabi Island (black), West Wallabi Island (shaded) and North Island (white).

Chapter 4 Island population genetics 106

4.3.2 mtDNA diversity

Within the 92 individuals examined, eight haplotypes were identified (Figure 4-4). Tammar wallabies from EWI had three haplotypes (A, B, and D) none of which were shared with WWI (haplotypes E, F, G and I). NI had the lowest number of haplotypes (F and H), one of which was shared with WWI (haplotype F), the other, haplotype H, was unique to NI (Figure 4-4). These control region haplotypes were distinguished by nine variable sites (Figure 4-5, Appendix 7). Higher haplotypic diversity and nucleotide diversity was detected in WWI and NI, than EWI (Table 4-4). Both S and

k were concordant, indicating that the effective population size (Ne), at each generation, is highest on WWI (S = 1.782; k = 1.016), followed by EWI (S = 0.985;

k = 0.592). NI (S = 0.505; k = 0.279) consistently had the lowest Ne.

1.0 0.9 0.8 0.7 0.6 0.5

Frequency 0.4 0.3 0.2 0.1 0.0 ABDEFGHI mtDNA haplotype

Figure 4-4 Distribution of mtDNA control region haplotypes across the three Abrolhos Island tammar wallaby (Macropus eugenii) populations, East Wallabi Island (black), West Wallabi Island (shaded) and North Island (white).

Chapter 4 Island population genetics 107

Table 4-4 Summary of mtDNA control region (642 bp) diversity indices for East Wallabi Island (EWI), West Wallabi Island (WWI) and North Island (NI) tammar wallaby (Macropus eugenii) populations.

EWI WWI NI n 33 29 30 No. haplotypes 3 4 2 h (± sd) 0.225 (± 0.092) 0.739 (± 0.039) 0.497 (± 0.042) (± sd) 0.352 (± 0.393) 2.759 (± 1.674) 0.993 (± 0.767) Sample size (n), haplotypic diversity (h) and nucleotide diversity ().

Chapter 4 Island population genetics 108 Variable site position 111111111 1111111111 1111112222 2222222222 2222222222 2222222233 12333344 5666777777 8888889999 9001223344 5556666777 8888990011 1233344444 4556666666 7789999901 Haplotype 1619056727 8347123478 2456780567 8487285823 2461257356 0135391612 4107834567 8170134679 0990136713 AID ?CTAAGCATG GACTTTATCT CTTCTCCAA- TGTTTCATGT AACATACCTA TCCCATTTTT TTTCTCACAT TATCCACATA AA-ATTTACG AIE ?...... - ...... -...... AIF ?...... - ...... -...... AIG ?...... - ...... -....G.. AIH ?...... - ...... -...... AII ?...... - ...... -...... AIA ?...... - ...... -...... AIB ?...... - ...... -..C.... TU7 T...... G...TTT ...... T....C...... -...... C. .G-..CCT.. TU8 T...... G...TTT ...... T....C...... -...... C. .G-..CCT.. TU9 T...... G.....- ..CC...... G.T...... C..T....C...... -..CC... TU10 T...... G.....- ..CC...... G.T...... C.TT....C...... -..CC... TU11 T...... G.....- ...C...... G...... T..C...... C. ..-..CC... TU12 T...... G.....- ...... T...T... C..T....C...... T...... -...C... PE23 T...... G.....- ...C...... GG....T...... C. ....C...... C. ..-..CC... PE24 T...... G.....- ...C...... T...... T.C. ..-..CC... K14 GT...... A...C.TC.. .A...... G ....CT.CAC G...C.AT.C CT.T..A..C .CAAA..... C..T.C.CAG CT-..CCTAA KI5 GT...... A...C.TC.. .A...... G ....CT.CA. G...C.AT.C CT.T..A..C .CAAA..... C..T.C.CAG CT-..CCTAA EGK ??AGGATC.C AGTC..TCTC TA.A.TT..- CA....GCA...... T.T.C .T.TG.-..C C.AAA.-ATA C.CTA..-C. C.T.AC.C.. WGK ??AG...CCC AG.C.C.CTC TGCACT...- .....A.C.. ..TG.C..AC ..A...-..C C.AAA.CATA CGCTA..-C. C.AGAC.T..

Chapter 4 Island population genetics 109

Variable site position 3333333333 3333333333 3333333333 3333333334 4444444444 4444444444 4444444444 4555555555 5555555555 5 1122222233 3444445555 5556666677 7788999990 0000111112 2222333334 5556677888 9001111222 2223444478 8 Haplotype 4802378912 3023790125 6785678903 4904124680 1379123480 5678014567 1250178278 9471247234 5672147975 6 AID CTTTCCTCAC AACATATCCG TAAAACTATC TCTACATCAA ACCCGTCCC- AAAATCACTT TATCCCCCTC ATCACAGCCC ACCACTCCCT C AIE ...... - ...... T...... AIF ...... C...... C...... T.....- ...... T...... AIG ...... T.....- .....T...... T...... AIH ...... T.....- ...... T...... AII ...... G...... T....T- ...... T...... AIA ...C...... T.....- ...... AIB ...C...... T.....- ...... TU7 ....T...... C.C..A ...... C...... T.A....- ...... C...... T.A...... TT. . TU8 ....T...... C.C..A ...... C...... T.A....- ...... C...... T.A...... TT. . TU9 ...... C.CT.A C...... C. C...... T...T..- ...... C..T...... A...... G.C.T.. . TU10 ...... C.CT.A C...... C. C...... T...... - ...... C..T...... A...... G.C.T.. . TU11 ....T...... C.T. C...... G.. C...... T..C...- ...... C. ..C...... A..T ...G.C.T.. . TU12 .C...... C.CTT. C...... C. C...... T...T..- ...... C..T...... A...... G.C.T.. . PE23 ...... C..A C....T..C. C.C..T...... T...... - ...... C..T...... A.T...... C.T.. . PE24 ....T...... C.T. C...... C. C...... C...- ...... C. *.A...T...... A..T ...G.C.TT. . K14 ..C.T..T.. ..T...C.AA A.G..AC.A. C.....CAT...... A...- ....C....A .....A.... C....CAT.. ...G.C.TT. . KI5 ..C.T..T.. ..T...C.AA A.G..AC.A. C.....CAT...... A...- ....C....A .....A.... C....CAT.. ...G.C.TT. . EGK ....TTCACT .GTGAGC.TC ..G.T.A.CA A..GT.CGGG TTT...ATAG G..G...T.C .T...A.A.T ..TT..AT.. CTTGT.T.TC T WGK T.C.TTCACT GCT...C.TA C.GGTTA.CA AT....CAGT ....A.A.AA .CC..TCT.. CT.TAA.ACT .CTT..A..T T..GTCT.TC T

Figure 4-5 Variable sites within the examined 595 base pair segment of mtDNA control region from 102 tammar wallabies (Macropus eugenii). Variable nucleotide positions are relative to the first sequence. Dots represent identical bases to the first haplotype, ? = missing data, - = indel. Tammar wallaby haplotypes are labeled relative to their location, that is, Abrolhos Islands (AI), mainland Western Australia (TU = Tutanning and PE = Perup), South Australian Kangaroo Island (KI); eastern grey kangaroos (EGK) and western grey kangaroos (WGK). *represents a 64 base pair indel sequence: (CAAGACCATAAACTCATACATTACTAAACTCAAATTTTACTAAATACATAGAATCAATGATAAA).

Chapter 4 Island population genetics 110

4.3.3 Population structure and gene flow

The STRUCTURE analysis posterior probabilities indicated the data were structured into two clusters (Figure 4-6(a)). The model value of K also suggested K = 2 (Figure 4-6 (b)). The assignment probabilities showed evidence of population structuring within the three island populations. There was no evidence of recent gene flow/admixture between EWI and WWI (Figure 4-7), while WWI and NI tammars were classed as a single population. There was significant divergence between populations compared with variation within populations (ST = 0.660, p = 0.00000). Fifty percent of the variation occurred between the two groups (EWI, and combined WWI and NI), 16% within the groups and 34% occurred within populations.

(a) (b) 0 4000

3000 -1000

K 2000

ln P(D) -2000 1000

-3000 0 0123456789 0123456789 No. of populations (K)

Figure 4-6 Posterior probabilities from the STRUCTURE analysis indicated the data were structured into two clusters. (a) The log probability data, Ln P(D), as a function of K; (b) The rate of change in the log probability of data, K, as a function of K.

Chapter 4 Island population genetics 111

East Wallabi Island West Wallabi Island North Island

Figure 4-7 The level of admixture among tammar wallabies (Macropus eugenii) in the three Wallabi Group Islands. Each vertical bar represents a single individual. Individuals were assigned to a population using the program STRUCTURE (Pritchard et al. 2000). The population number, K, was estimated using no prior population information.

4.3.4 Phylogenetics

The reconstruction of the relationships amongst EWI, WWI and NI mtDNA haplotypes using NJ, MP and ML methods produced trees with similar topologies and bootstrap support for the resolved groups (Figure 4-8). The ML analysis resulted in a phylogenetic tree that distinguished four main groups (Abrolhos, mainland WA, SA (KI) and the grey kangaroos) all with strong bootstrap support ( 96% in ML) and Bayesian credibility values ( 0.70). Within tammars, there was strong support (100%) for two major clades consisting of tammars from WA and SA respectively. There was also strong (96%) support for the mtDNA from Abrolhos Island tammars being monophyletic with respect to mainland WA tammars, but they were not reciprocally monophyletic. The haplotypes from mainland WA tammars generally exhibited an unresolved polytomy that was resolved in MP trees. Sequence divergence (mean ± se) within the Abrolhos was 0.537 ± 0.073% and between the Abrolhos and mainland WA was 4.298 ± 0.061%. The largest divergence was between the haplotypes from Abrolhos and SA (11.319 ± 0.046%). As expected, the sequence divergence between the tammar wallaby and grey kangaroo haplotypes was also high (17.932 ± 0.102%).

Chapter 4 Island population genetics 112

AID

AIF

AIH

96 AIE 0.76 Abrolhos AII Islands, WA

AIG

57 AIB

0.90 AIA

PE23 100 59 TU9WA9 1.00 0.70 87 TU12WA12 1.00 Mainland, TU10WA10 WA PE24 100 80 1.00 TU11 1.00 WA11

100 WA8TU8

1.00 WA7TU7

100 KI4 Kangaroo Island, SA 1.00 KI5

WGK

EGK 10

Figure 4-8 Maximum likelihood (ML) tree of relationship amongst Abrolhos, mainland WA and SA tammar wallaby (Macropus eugenii) mtDNA control region sequences (595 bp). Robustness is indicated by bootstrap values ( 50%, above branches) and Bayesian posterior probabilities ( 0.50, below branches).

Chapter 4 Island population genetics 113

4.3.5 Computer simulations

Estimates of Ne were calculated from the autosomal microsatellite data using

ONeSAMP (Tallmon et al. 2008) for EWI (mean Ne = 24, 95% confidence interval

(CI) = 18.5 – 34.1) and WWI (mean Ne = 23, 95% CI = 23.1 – 47.6). Following the islands’ isolation from the mainland, the rate of loss of genetic diversity due to drift within EWI and WWI populations was predicted using the simulation program GENELOSS (England & Osler 2001). Under the influence of genetic drift alone, all variation on both East and West Wallabi would be lost within 200 to 300 generations

(AD = 1.00; He = 0.00) when Ne = 40 and Ne = 60. When Ne = 100, diversity was maintained for a longer period, but was still lost within 400 to 500 generations (Figure 4-9).

(a)

12 1.0 ) e

10 H 0.8 8 0.6 6 4 0.4 2 0.2 Allelic diversity(AD) 0 ( Heterozygosity 0.0 0 100 200 300 400 500 600 700 800 0 100 200 300 400 500 600 700 800

Generations

(b)

12 1.0 ) 10 e H 0.8 8 0.6 6 4 0.4 2 0.2 Allelic diversity (AD) 0 ( Heterozygosity 0.0 0 100 200 300 400 500 600 700 800 0 100 200 300 400 500 600 700 800 Generations

Figure 4-9 The predicted rate of loss of genetic diversity (allelic diversity, AD and heterozygosity, He) for (a) East Wallabi Island and (b) West Wallabi Island populations of tammar wallabies (Macropus eugenii) in the Abrolhos Archipelago (Ne

= 40 (); Ne = 60 (); Ne = 100 ()).

Chapter 4 Island population genetics 114

4.4 Discussion

This is the first investigation into the genetics of tammar wallaby populations in the Houtman Abrolhos Archipelago. The autosomal microsatellite, Y-linked microsatellite and mtDNA data revealed that genetic diversity was highest on the largest island (WWI), and lowest on the smallest (NI). All three island populations were inbred and had significantly lower levels of genetic variation than their WA mainland counterpart. The three Abrolhos populations were significantly differentiated from one another, as well as the mainland population. There was no evidence of recent gene flow between EWI and WWI, despite their geographic proximity as each island had a high proportion of private microsatellite alleles, as well as unique mtDNA and Y- chromosome haplotypes. The microsatellite and mtDNA data indicate that the NI population is a subset of WWI, contrary to local belief that they were founded by individuals sourced from EWI. The low levels of genetic diversity detected on NI are consistent with the reported small number of founders. Two main lineages were detected within tammar wallabies (SA and WA) following phylogenetic analysis of the mtDNA control region sequence data. Within the WA clade, the Abrolhos tammar wallabies were monophyletic suggesting that they should be considered as a separate MU.

4.4.1 Genetic diversity

The genetic diversity in the Abrolhos tammar wallaby populations was significantly lower than the adjacent WA mainland population, reflecting their long term isolation and limited population size. Within the Abrolhos, the largest island (WWI) had the highest microsatellite and haplotypic diversity, and the smallest island (NI), had the lowest. These data match theoretical predictions that (i) genetic variation within a species is related to population size; (ii) genetic variation is related to island size; and (iii) island populations have less genetic variation than mainland populations (Frankham 1997; Soulé 1976). Since no census data are available, it is assumed that the larger island supports a larger tammar population since more habitat is available.

Studies examining other island tammar populations have shown similar results. Eldridge et al. (2004) compared the genetic diversity of Garden Island (WA) and

Chapter 4 Island population genetics 115

Tutanning (WA) tammars, and detected similar trends in heterozygosity (Garden

Island He = 0.44; Tutanning He = 0.86) and allelic diversity (Garden Island AD = 3.7;

Tutanning AD = 13.9) as this study (mean Abrolhos He = 0.42, AD = 3.5; Tutanning He = 0.82, AD = 10; Table 4-1) despite genotyping individuals for different microsatellite loci. A similar pattern of significantly reduced diversity in island compared to mainland populations has also been found for other Australian marsupials, for example dibblers (Parantechinus apicalis) (Mills et al. 2004); quokkas (Setonix brachyurus) (Sinclair 2001); eastern barred bandicoots (Perameles gunnii) (Robinson et al. 1993); rufous hare-wallabies (Lagorchestes hirsutus) and black-footed rock wallabies (Petrogale lateralis) (Eldridge et al. 2004), as well as a variety of other taxa (Frankham 1997). This demonstrates that while the island populations may have been shielded from threatening processes occurring on the mainland, such as predation and habitat loss, they have nonetheless been considerably affected by the genetic and demographic processes intrinsic to island populations, such as loss of genetic diversity through isolation and genetic drift, making them inherently vulnerable to extinction.

Significant levels of effective inbreeding were detected within each island population, but there was no evidence from the microsatellite data of a recent genetic bottleneck. The inbreeding detected has arisen as a consequence of small population size and long- term isolation and is associated with a high prevalence of kinked tails 53.8% (35 of 65) in animals on EWI and WWI (Plate 4-2). Kinked tails were also commonly observed on NI, although the frequency was not recorded. Additionally, on NI a small proportion of males 3.4% (3 of 89) exhibited unilateral cryptorchidism, whereby one testis had not descended properly (Plate 4-3). Morphological abnormalities, unilateral cryptorchidism and associated sperm abnormalities, have also been observed in some introduced island koala (Phascolarctos cinereus) populations (Seymour et al. 2001). These traits, which are known to be associated with inbreeding (Frankham et al. 2002), can result in reduced fitness and reproductive success. Kinked tails, cowlicks, cryptorchidism, poor sperm viability and sterility have been documented in the Florida panther (Felis concolor coryi) a subspecies of the North American mountain lion (also know as the puma or cougar) that is suffering from reduced fitness due to fixation of detrimental alleles, and low genetic diversity (Hedrick 1995). To restore genetic variation and remove some of the detrimental alleles in the Florida panther population,

Chapter 4 Island population genetics 116

Texas cougars (Felis concolor stanleyana), a related subspecies, were translocated into the population (Hedrick 1995), and the resultant offspring have shown a reduced occurrence of kinked tails, cowlicks and cryptorchidism (Land et al. 2004). Augmenting the Abrolhos populations with mainland WA wallabies may similarly help overcome any deleterious traits associated with inbreeding.

(a)

(b)

Plate 4-2 Evidence of kinked tails in the Abrolhos Island populations of tammar wallabies, Macropus eugenii (a) categorised subjectively from right to left, a major kink, a minor kink and a straight tail, (b) close up of a tail with a more prominent kink.

Chapter 4 Island population genetics 117

Plate 4-3 Photographs of sexually mature male tammar wallabies (Macropus eugenii) exhibiting (a) unilateral cryptorchism on North Island, Western Australia, and (b) a male with normal testicles from Kangaroo Island, South Australia.

Chapter 4 Island population genetics 118

Although the NI population is believed to have been founded by a small number of individuals and is inbred, the population continued to flourish in a relatively benign environment where resources are abundant. In such circumstances, even species with reduced fitness are likely to be successful, for example introduced koala (P. cinereus) populations in South Australia (Seymour et al. 2001). If a bottleneck occurred at the introduction of tammars to North Island, then it appears to have been more demographic than genetic with the population increasing rapidly to a large size. Therefore only limited loss of genetic diversity occurred (Frankham et al. 2002).

The detrimental effects of low genetic diversity and genetic drift may be exacerbated by low (or absent in this case) gene flow. Under the influence of genetic drift, the

GENELOSS simulations suggest that even when Ne = 100, there should be no genetic variation persisting in the EWI and WWI populations after 400 – 500 generations. The generation time for tammar wallabies is two to three years, and the Abrolhos Islands have been isolated from the mainland for at least 4600 generations. EWI and WWI have been separated from one another for at least 2600 generations, yet these populations have maintained some level of genetic variation. Mutation counters genetic drift and adds variation back into populations. Using the step-wise mutation model (Slatkin 1995), we calculated the expected heterozygosity for EWI and WWI for drift-mutation equilibrium, given as:

0.5 He = 1 – 1 / (1 + 8 Ne)

(Ohta & Kimura 1973). When Ne = 24 (see Results section 4.4.5) and the mutation rate -3 () for microsatellites is 10 (Weber & Wong 1993), the He for EWI tammar wallabies -3 is 0.161. When Ne = 23 and = 10 , the He for WWI is 0.155. The observed values for heterozygosity (Table 4-1) are higher than the drift mutation equilibrium values, therefore there is more variation present in these populations than we expect under drift-mutation equilibrium. Genetic variation is generated through mutation, selection and immigration, however it is unlikely that these explain the higher He observed. First, microsatellite mutation rates for mammals have been estimated at 10-2 to 10-5 (Dallas 1992; Ellegren 1995; Weber & Wong 1993) and the mutation rate used in this study was within this range. Second, microsatellites are selectively neutral (Li et al.

Chapter 4 Island population genetics 119

2002), and third, the genetic data shows no evidence of immigration. The most likely explanation is that the Ne for the Abrolhos tammar populations has been underestimated. Although the estimates were consistent when altering parameters in the ONeSAMP software, the 95% confidence intervals around Ne were large. However a tripling of Ne (~ 69) was required to obtain the value of 0.356 estimated under the drift-mutation equilibrium model. This value falls outside of the 95% CI estimates calculated from the microsatellite data using ONeSAMP. Estimating Ne is complex as it is influenced by several factors such as genetic drift, asymmetric gene flow, and linkage disequilibrium (Fraser et al. 2007) although temporal sampling has been recommended to improve estimates of Ne (Schwartz et al. 1998). Therefore census data for these populations is needed and when combined with the microsatellite data, may provide better Ne estimates.

4.4.2 Population structure and gene flow

As there is no evidence for recent gene flow amongst populations within the Abrolhos, it was not surprising that all three populations showed significant genetic differentiation. This was supported by evidence from both the microsatellite and the mtDNA data. Within the Abrolhos, EWI and WWI had the highest level of differentiation, and WWI and NI showed a lower degree of differentiation. The low levels of admixture between EWI and WWI indicates there has been no recent gene flow and each island has been evolving independently since their separation 6 000 years ago. Gene flow prevents genetic differentiation between populations and maintains genetic diversity. Without gene flow, each population will become genetically distinct due to other factors such as genetic drift (Allendorf & Luikart 2007). This result was unexpected as EWI and WWI are separated by only 1.9km of shallow ocean, much of which is exposed during low tide (BOM 2008). From this we can conclude that tammar wallabies are poor dispersers over water. Given the isolation and the absence of gene flow, these Abrolhos Island populations are expected to continue to lose variation due to genetic drift and diverge from one another, and the mainland.

Chapter 4 Island population genetics 120

4.4.3 North Island founder and population genetics

A surprising result from this study was the relationship between tammars from WWI and NI. Despite the local belief that NI tammars originated from EWI, the genetic data indicate that the NI population are clearly derived from the population on WWI. NI tammars shared 100% of their autosomal and Y-linked microsatellite alleles, as well as one mtDNA haplotype with WWI and not EWI. The second mtDNA haplotype found on NI (haplotype H) is most closely related to WWI haplotype G (one base pair difference) and so most likely represents a recent mutation on NI or a haplotype that occurs at low frequency on WWI and was not detected.

If tammars do not disperse readily across the narrow sheltered gap (1.9km) between EWI and WWI, it is inconceivable that they were able to disperse 14km north-west through the open ocean to reach NI from WWI. Therefore the NI population has most likely been deliberately introduced. Although local oral history records two introductions of tammars to NI from EWI in the 1980s, the current population is clearly of WWI origin. So either the oral history is inaccurate or the animals introduced from EWI failed to establish and there has been a subsequent undocumented but highly successful introduction of animals from WWI to NI. It is possible that some animals of EWI origin or descent do exist on NI but at such low frequencies that they were not sampled. However, this is unlikely since over 100 NI individuals have now been genotyped (unpublished data) and no microsatellite alleles, or mtDNA haplotypes characteristic of EWI have been detected.

The NI tammar population shows evidence of a limited number of founders since it has significantly lower AD compared to its source population (WWI) and is monomorphic at one locus. As predicted by theory and demonstrated by these data, allelic diversity is more sensitive to founder effects than heterozygosity (Allendorf 1986; Nei et al. 1975). Establishing populations with a small number of individuals can result in a genetic bottleneck, changes in allele frequencies and a loss of diversity because often the source individuals represent only a fraction of the genetic variants from the source population (Nei et al. 1975). The negative impacts of small founder numbers and genetic bottlenecks have been documented in vertebrates (Hedrick et al. 2001b; Jones et al. 2004; Ramstad et al. 2004; Tarr et al. 1998), invertebrates (Johnson 1988) and

Chapter 4 Island population genetics 121 invasive species that have undergone sequential founding events (Cabe 1998; Dlugosch & Parker 2008; Ficetola et al. 2008). However no genetic signature of a bottleneck was detected in the NI population perhaps due to a Type II error (failing to detect a bottleneck when there was one), or possibly the population had made a rapid demographic recovery. Detecting bottlenecks can also be difficult when using single samples opposed to temporally spaced samples (Cornuet & Luikart 1996; Williamson- Natesan 2005).

The isolation of island populations exacerbates the effects of small numbers of founders though the associated effects of small populations size, restricted gene flow, random genetic drift, genetic bottlenecks and inbreeding (Eldridge et al. 1999; Frankham 1997, 1998). Since islands are increasingly being used as refuges for threatened species, and a source of animals for conservation initiatives (Boessenkool et al. 2007; Courtenay & Friend 2003; Daltry et al. 2001; Eldridge et al. 2004; Rankmore et al. 2008), it is fundamental for conservation planning to take into consideration the number of founders and genetic variation for long-term planning. Establishing colonies on islands may protect them from deterministic threats on the mainland, but will increase the impact of stochastic events such as random genetic drift, particularly if they are founded with a small number of individuals that may already be inbred due to population decline on the mainland. In these situations it would be beneficial to regularly exchange individuals between the introduced island and existing mainland populations to simulate natural gene flow and so assist in the maintenance of genetic diversity.

4.4.4 Phylogenetics and conservation significance

The phylogenetic analysis of Abrolhos, mainland WA and SA tammar wallaby mtDNA showed reciprocal monophyly between the WA and SA tammar lineages, which was well supported by bootstrap values. Within the WA clade, the Abrolhos tammar wallaby haplotypes were monophyletic. Although the mainland WA haplotypes are clustered together they do not exhibit complete reciprocal monophyly with the Abrolhos haplotypes. On the basis of these data WA and SA tammar wallaby populations should be classified as separate ESUs on the basis of reciprocal monophyly at mtDNA control region (Moritz 1994; Moritz 1999; Waples 1991). MUs

Chapter 4 Island population genetics 122 are identified by significant differences in allele frequency distributions and significant divergence in their mtDNA, therefore the mainland WA, EWI and WWI/NI warrant being treated as separate MUs. The high level of divergence between WA and SA tammar wallabies warrants further analyses and may result in taxonomic revision. However any future analysis should include all extant tammar wallaby populations.

Existing data indicate that the tammar wallaby populations in SA and WA represent two major reservoirs of diversity within M. eugenii, and both should be conserved as they reflect the long term historical isolation of populations within the species. The separate MUs including EWI, and WWI/NI, and mainland WA should be managed in such a way as to maintain diversity within the WA tammar wallaby ESU. There are several options for how the Abrolhos tammar populations could be managed. NI is an introduced population that has become a pest species on the island and was subject to a culling operation in 2008. The loss of the NI population through extinction or eradication would not be catastrophic as the diversity present on NI is present in the WWI population. However, if either EWI or WWI were to go extinct it would represent the irreplaceable loss of unique populations that are significantly different from one another, and all other tammar populations. Since island populations are known to be prone to extinction (Frankham 1998), NI could provide a potential ‘backup’ for WWI. However the current NI population is not particularly representative of the WWI population as they were founded by a small number of individuals, have lower genetic diversity and higher levels of inbreeding compared to the other Abrolhos populations.

Since all the Abrolhos tammar populations are inbred, subdivided and have no gene flow, they may benefit from augmentation, that is, translocation of individuals from the larger mainland population into the smaller island populations (Moritz 1999). From a demographic and genetic perspective, the Abrolhos populations would benefit by increasing genetic diversity, reducing the impact of inbreeding and preventing further genetic deterioration (e.g., kinked tails). But is this appropriate? The Abrolhos tammar wallabies are morphologically and reproductively distinct from the mainland (both WA and SA) populations (Herbert et al. unpublished data). The environmental conditions on the Abrolhos Islands are so harsh that a reciprocal augmentation may be

Chapter 4 Island population genetics 123 unsuccessful as mainland tammar wallabies may be unable to survive, particularly with the potential future effects of climate change on the environment. Furthermore, it is not recommended that inbred or genetically depauperate populations be used to augment other populations in an effort to maximise genetic diversity as it can compromise their long-term evolutionary plasticity (Eldridge et al. 1999; Moritz 1999). Augmenting the Abrolhos tammar populations with mainland WA tammars also raises the issue of whether these island populations should be allowed to retain their genetic integrity or be introgressed to reduce their risk of extinction. Alternatively, NI could be augmented with individuals from WWI and EWI, transforming NI into a reservoir for a generic Abrolhos tammar population and thus maintaining the genetic integrity of both EWI and WWI.

4.4.5 Conclusions

This study examined the population genetics of tammar wallabies from three islands in the Houtman Abrolhos Archipelago, WA. Consistent with population genetic theory, there was a relationship between island size, population size and genetic diversity. WWI had the highest levels of allelic and haplotypic diversity, followed by EWI and then NI. The Abrolhos populations have significantly lower genetic diversity than mainland tammars from Tutanning Nature Reserve and all sampled populations are significantly differentiated. The significant genetic differentiation between the EWI and WWI tammar populations suggests no recent gene flow between the two islands in spite of their relatively close proximity. Isolation and lack of recent gene flow have led to significant genetic differentiation within the Abrolhos, as well as between the Abrolhos and the mainland tammar wallaby population. One of the biggest surprises of this study was the revelation that the NI tammar population was founded by animals from WWI, which does not concord with local beliefs. The mtDNA data indicates that SA and WA tammar wallabies should be considered as separate ESUs, and the Abrolhos and mainland populations be treated as separate MUs. This investigation highlights the importance of incorporating genetic strategies into management plans when considering utilising islands as refuges for mainland populations as they are prone to the risk of extinction, loss of genetic diversity and inbreeding depressions.

Chapter 5 Traits influencing male reproductive success 124

Chapter 5 Dominance, body size and internal relatedness influence male reproductive success in eastern grey kangaroos (Macropus giganteus)

5.1 Introduction

Reproductive success is defined as the number of surviving offspring produced by an individual (Clutton-Brock 1988). Research into the variance of reproductive success in males and females is essential for understanding the role of ecology, demography and genetic structure in the process of natural selection (Clutton-Brock 1988). In most mammals with no parental care, males maximise their fitness by producing as many offspring as possible (Emlen & Oring 1977; Trivers 1972). Consequently, males compete intensely for access to females, and this contributes to the variance in reproductive success among males (Andersson 1994). Male reproductive success is an outcome of several interacting processes. On a pre-copulatory level, individual behaviours such as fighting ability, mate guarding and weaponry can alter the probability of a male gaining a successful mating. On a post-copulatory level, when several males copulate with the same female, the sperm of each male competes for fertilisation (Birkhead & Møller 1998). Cryptic female mate choice can manipulate the outcome of paternity through selection for a particular male’s sperm within the female reproductive tract (Eberhard 1996). Since reproductive success is not directly observable, other measures such as male dominance status, body size, and mating success have been used as surrogate measures. However, this is problematic as several studies have shown mating success does not always reflect paternity success (Coltman et al. 1999a; Hughes 1998; Issac 2005).

Chapter 5 Traits influencing male reproductive success 125

In the priority-of-access model Altmann (1962) asserts that in species with a hierarchical social system, highly ranked males should monopolise mating when there is at least one receptive female. There is much evidence demonstrating that priority of access to females is frequently an outcome of social dominance (Dewsbury 1982), and empirical data has shown in polygynous species, dominance is often correlated with male reproductive success (Ellis 1995; Engelhardt et al. 2006; Røed et al. 2002; Spong et al. 2008). ‘Dominance’ can be broadly defined as success in competition or contests. Dominant males can exclude at least some of their rivals from access to mates or resources critical for attracting mates. This is achieved by driving away, killing or by other form(s) of intimidation. There is evidence that traits such as larger body size, aggressiveness, weaponry size or signals of fighting ability (badges of status) assist in the establishment of dominance hierarchies (Qvarnström & Forsgren 1998). However, these behavioural and morphological traits can be costly, for example an increased risk of disease susceptibility, predation and energy stress (Zahavi 1975). Therefore, if only males of relatively high quality are able to bear the cost of dominance, the traits indicating dominance and an individual’s position in a hierarchy per se are thought to be a reliable indicator of certain aspects of male quality. However, dominance hierarchies can be unstable. How an individual values the contested resource varies and as a result the best fighter may not necessarily always win a contest (Beacham 2003).

In mating systems in which multiple males compete for access to females, it can be predicted that selection will favour phenotypic adaptations that enhance a male’s ability to gain access to females or resources that attract females (Schulte-Hostedde et al. 2001b). Positive relationships have been found between large body size and dominance rank in some studies (Charpentier et al. 2005; Clark & Faulkes 1998; Pelletier & Festa-Bianchet 2006; Schuett 1997), but not others (Berard et al. 1993; Henderson & Hart 1995; Marvan et al. 2006). Additionally, large body size has often been shown to enhance a male’s reproductive success, for example the common brushtail possum (Trichosurus vulpecula) (Clinchy et al. 2004), bridled nailtail wallaby (Onychogalea fraenata) (Fisher & Lara 1999), brown antechinus (Antechinus stuartii) (Fisher & Cockburn 2005), but this is not always the case (Coltman et al. 1999b). Other factors such as age and/or weaponry can also be important determinants

Chapter 5 Traits influencing male reproductive success 126

(Clutton-Brock et al. 1982; Coltman et al. 2001; Henderson & Hart 1995; Rasmussen et al. 2008; Weatherhead et al. 2002). The revolution in molecular technology has shed light on cryptic alternative mating tactics that are less dependant on body size such as ‘sneaker’ strategy males and sperm competition (Coltman et al. 2001; Preston et al. 2001).

The steroid hormone testosterone is linked to reproduction in vertebrates, via its influence on morphological, physiological and behavioural traits such as secondary sexual characters (Wickings & Dixson 1992; Wingfield et al. 1990), male sexual physiology and behaviour (Dixson 1998; Rudd et al. 1996), as well as being often associated with social status and aggression (Bergman et al. 2006; Dixson 1998). Since dominance often results in a higher reproductive success, it is logical that the dominant male would have higher testosterone concentrations. Indeed some studies have found a correlation between dominance rank and testosterone (Clark & Faulkes 1998; Muller & Wrangham 2004; Perret 1992; Setchell & Dixson 2001; Setchell et al. 2008), while other studies have not (Barrett et al. 2002; Carlson et al. 2004; Hynes et al. 2005; Klinkova et al. 2004). Testosterone has also been linked directly to an increase in reproductive success for some males (O'Neal et al. 2008), or it may indirectly influence success through the enhancement of secondary sexually selected traits.

Female mate choice may also influence variance in male reproductive success (Andersson 1994; Clutton-Brock 1989) through the selection of characteristics in males other than secondary sexually selected traits. Females invest more in their offspring than males, leading to the prediction that they should be more choosy and select mates of high genetic quality (Trivers 1972). It has long been established across a wide range of taxa, that mating between close relatives (inbreeding) causes a decline in reproductive fitness as a result of an increase in homozygosity (inbreeding depression) (Keller & Waller 2002). Avoidance of consanguineous matings is thought to influence the evolution of reproductive behaviours (Clutton-Brock 1989) and thus select for genetically different individuals. Superior male competitors may have higher genetic diversity (heterozygosity) and there is growing evidence of a relationship between levels of individual heterozygosity and fitness traits, for example reduced susceptibility to diseases and parasites (Acevedo-Whitehouse et al. 2003); territory

Chapter 5 Traits influencing male reproductive success 127 size, song structure and reproductive success (Seddon et al. 2004); and increased survivorship (Coltman et al. 1998). Furthermore, reproductive success has been shown to be negatively correlated with parental similarity (Amos et al. 2001a; Hoffman et al. 2004).

Marsupials are a model species for research as they provide important evolutionary information about mammals and represent the dominant mammalian group in Australia. Few studies have examined what traits potentially influence reproductive success in marsupials, and more specifically macropodids (kangaroos and wallabies). Among macropodids, several studies have shown a positive relationship between body size and dominance, in that the largest male is the most dominant (the alpha male) including in the red kangaroo (Macropus rufus) (Croft 1981; Russell 1970), whiptail wallaby (M. parryi) (Kaufman 1974), tammar wallaby (M. eugenii) (Hynes et al. 2005; Rudd 1994), red-necked wallaby (M. rufogriseus) (Johnson 1989) and bridled nailtail wallaby (Fisher & Lara 1999). Studies examining the relationship between reproductive success and male dominance in the tammar wallaby have produced conflicting results. Two studies found that the dominant male sired the most offspring (Hynes et al. 2005; Miller et al. in press), whereas a third study did not (Ewen et al. 1993). Several behavioural studies conducted in the eastern grey kangaroo (M. giganteus) state that the most dominant male has the highest reproductive success (Jarman 1983, 1989; Jarman 1991; Jarman & Southwell 1986), although no molecular data supports this assertion.

The eastern grey kangaroo (Plate 5-1) is an excellent model to examine a broad range of traits potentially influencing male reproductive success. It is a polygynous species with a complex social organisation and displays extreme sexual size dimorphism (Jarman 1983). Eastern grey kangaroos are gregarious, territorial and hierarchical. Males compete intensively for access to resources, including females, through a range of agonistic behaviours (Ganslosser 1989; Jarman & Southwell 1986). Eastern grey kangaroos live in small groups of less than six individuals that consist mainly of mothers and daughters. At dawn and dusk, several groups whose home-ranges overlap, aggregate where there are concentrated resources, for example a preferred feeding area, forming what is called a mob. Young adults, generally males, may disperse to

Chapter 5 Traits influencing male reproductive success 128 enter other mobs (Kaufmann 1974). During winter, when females are less likely to come into oestrus, dominant males have been observed to separate themselves from the main mobs (Kaufmann 1974). Observational data alone cannot be relied upon to accurately estimate male reproductive success and long-term genetic studies are essential to clarify the relationship between sexual size dimorphism, mating success and paternity in this and other species.

The aim of this study was to investigate the relative importance of traits potentially influencing male reproductive success in a polygynous marsupial using the eastern grey kangaroo as a model species. More specifically, we investigated whether male reproductive success was influenced by: (i) dominance status; (ii) morphological traits such as body weight and body size; (iii) testosterone concentrations; and (iv) genetic traits such as internal relatedness, standardised heterozygosity and mean d2, the latter two being measures of genetic diversity. Despite the wealth of knowledge on the social organisation of kangaroos, this is the first study to examine parentage and male reproductive success, as well as its influence on their mating system.

Chapter 5 Traits influencing male reproductive success 129

Plate 5-1 Male eastern grey kangaroos (Macropus giganteus).

5.2 Materials and Methods

5.2.1 Study populations

This study was conducted in three semi free-ranging captive populations of eastern grey kangaroo in New South Wales, Australia (Table 5-1). Population A was located at the University of New South Wales Research Facility (33° 55’S, 150° 70’E). This colony was enclosed within nine hectares (ha) of natural bushland that was one third dry sclerophyll forest, and two thirds open grassland. Population B was located in a 12ha predator proof sanctuary of natural bushland (33° 40’S 151° 13’E) with two thirds of the area forested and one third grassland. Population C was located north of Sydney (33° 19’S 151° 14’E) in 68ha predator proof sanctuary of natural bushland.

Chapter 5 Traits influencing male reproductive success 130

Table 5-1 Number of eastern grey kangaroos (Macropus giganteus) observed (sampled and genotyped) in each of the three semi-free ranging populations.

Population Adult males Adult females Young-at-foot Pouch Young A 7 22 10 13 B 9 7 4 4 C 6 7 6 3

5.2.2 Animal capture and handling

Animal capture took place early in the morning, from 0600 in summer and 0730 in winter, and not during inclement weather. Kangaroos were captured by chemical restraint using a pole syringe or a blow pipe to deliver tiletamine hydrochloric acid (HCl) in combination with zolazapam HCl (Zoletil 100®, approximately 5mg/kg IM; Virbac Australia Pty. Ltd., Peakhurst, NSW). Animals that could be approached within 3m were immobilised with the aid of a pole syringe. At greater distances (up to 25m) a blow pipe was used to fire a dart (1CC, ¾”, volume 1ml, PA 17703, Pneu-Dart Inc., Williamsport) containing the anaesthetic agent into the thigh muscle of kangaroos.

All animals were weighed using 100kg Salter suspended dial scales (SA-235/9, Wedderburn, accuracy 0.05kg). While restrained in a jute sack, a range of morphological measurements were collected from adult males using vernier calipers or a tape measure (Plate 5-2). Skeletal measurements (head, leg, tail, pes (foot) and forearm length) and muscle circumference (tail base, forearm and bicep) were collected as indicators of male body size. Testis length and testis breadth were measured and used to calculate testicular volume as a large testes size relative to body mass can be indicative of sperm competition (Kenagy & Trombulak 1986; Short 1979). All testicular measurements exclude the epididymis and the same testis was always used for length and breadth. Testicular volume was calculated using the formula for an oblate spheroid, V = (/6).B2.L, where V = volume, B = breadth of testis and L = length of testis (Williamson et al. 1990). All measurements were taken by the same individual (E.J.M) to limit variability.

Chapter 5 Traits influencing male reproductive success 131

(a) (b)

(e) (f)

Plate 5-2 Collecting a range of morphological measurements (cm) from male eastern grey kangaroos (Macropus giganteus). (a) head length, (b) forearm length, (c) leg length, (d) pes length, (e) tail length, and (f) testis size.

Chapter 5 Traits influencing male reproductive success 132

Blood was collected from the lateral caudal vein of animals using a 21 gauge winged infusion set (Surflo, Terumo Corporation, Japan) and a 5ml syringe. Blood was transferred immediately into serum separation tubes (Vacuette, Greiner Labortechnik, Austria). The serum was separated and stored in duplicate aliquots at -20 C until assayed to determine the concentrations of testosterone.

The pouches of females were checked for the presence of young during each quarterly capture. The sex of all pouch young (PY) was recorded and the head, leg, and pes length was measured to the nearest 0.1mm using digital vernier calipers. PY were subsequently removed and euthanased as part of ongoing population management and a small tissue sample was collected from each for DNA extraction.

5.2.3 Behavioural dominance

At first capture, all adult males were ear tagged with a unique left/right ear colour combination (Allflex round button ear tags, Allflex Australia) for observational purposes. Ad libitum behavioural observations were carried out by a single observer (E.J.M.) during 2005 and 2006. Observations were conducted at dawn and/or dusk using binoculars (Pentax 10 x 50 PCF WP II). Each population was observed for an average of 130 (± 12) hours. All agonistic behaviours (Ganslosser 1989) initiated by or toward another male were recorded and scored as a win or loss (Martin & Bateson 1993).

5.2.4 Genetic assignment of paternity

DNA was extracted from small (2mm diameter) ear biopsies using standard methods (Sunnucks & Hales 1996; Appendix 1). Each individual (n = 98) was genotyped at 10 fluorescently labeled microsatellite loci characterised from macropod species (eastern grey kangaroo G16-1, G26-4, G31-1 and G20-2; tammar wallaby T19-1, T46-5, T31- 1, T32-1 and T3-1T; allied rock wallaby (Petrogale assimilis) Pa595) (Spencer et al. 1995; Zenger & Cooper 2001; Zenger et al. 2003). Genotyping was carried out using multiplexed PCR. Three combinations were used due to the similar allele size range in many of the loci in the eastern grey kangaroo: (i) T19-1, G16-1 and G26-4; (ii) G31-1, T46-5 and Pa595; and (iii) T31-1, G20-2, T32-1 and T3-1T. All loci were best

Chapter 5 Traits influencing male reproductive success 133 amplified in a “touchdown” PCR, whereby the initial annealing temperature was decreased at 3 – 4ºC increments and run for a total of 10 cycles each. Ten microlitre (μl) reactions were performed using a Multiplex PCR kit (Qiagen, Australia), 0.2μM of each primer and 30ng genomic DNA. Amplifications were carried out in a PTC-220 thermocycler (MJ Research, USA) using an initial HotStarTaq activation step at 95ºC for 15 min, a total of 50 cycles of 94ºC for 30 s, 64, 60, 56, 53 and 50ºC for 1 min 30 s (10 cycles each), 72ºC for 1 min 30 s, and final extension at 72ºC for 10 min. The PCR products were analysed in a 48 capillary AB 3730 DNA Analyser (Applied Biosystems, USA). The DNA fragments were sized and quantified using GeneMapper 3.7 (Applied Biosystems, USA).

5.2.5 Paternity analysis

The genotype file was checked for duplicate entries, scoring errors due to null alleles, short allele dominance and stutter bands using the program MICRO-CHECKER (van Oosterhout et al. 2004). Following genotyping, the data were checked to ensure Mendelian inheritance between mothers and their offspring. Since maternity was known, paternally derived alleles were readily identified. Paternity was determined by total exclusion manually, since each PY could only have been sired by one of the males present within the population. Males were assigned paternity if they possessed the identified paternal alleles at all 10 loci.

5.2.6 Testosterone radioimmunoassay

Testosterone concentrations were determined using the automated Immulite® 2000 immunoassay analyser with a solid phase competitive chemiluminescent enzyme immunoassay. Single 20l serum samples were assayed using the Immulite® total testosterone kit (L2KTW2, Diagnostics Product Corporation, Los Angeles, USA). The solid phase is made up of a polystyrene bead enclosed within the Immulite test unit that is coated with a polyclonal rabbit antibody for testosterone. The antibody is highly specific for testosterone, with a cross-reactivity of 0.6% with androstenedione, 0.5% with 5-androstan-3, 17-diol, 2% with 5-dihydro-testsoterone, 0.7% with methyltestosterone and 0.1% with progesterone. All samples were run in the same assay to reduce variability. This assay has been validated for use in a closely related

Chapter 5 Traits influencing male reproductive success 134 species, the tammar wallaby, in which serial dilutions of male tammar wallaby serum in charcoal stripped wallaby serum were found to be linear and parallel to serial dilutions of human plasma (Herbert et al. 2007). The assay sensitivity was 0.2ng/ml and the intra-assay coefficient of variation for three (low, medium and high) quality control pools containing 5.32 ± 0.17, 9.45 ± 1.11, and 19.80 ± 1.65 nmol/L was 3.24%, 11.80% and 8.34% respectively.

5.2.7 Data analysis

Agonistic interactions between males were arranged into a dominance matrix (Martin

& Bateson 1993). For each population, a dominance index (I) = Na/(Na + Nb) was calculated for each male, where Na is the number of encounters won and Nb is the number of encounters lost (Appendix 8). Males with the highest I values (> 0.7) were 3 considered dominant (Say et al. 2001). Landau’s linearity index (h) = 12/(n – n)* (va – ½ (n – 1))2 was used to estimate the linearity of the dominance hierarchy, where n is the number of animals and va is the number of individuals whom a has dominated. The h values range from zero (non-linear) to 1.0 (perfect linearity), with values > 0.9 considered significantly linear (Martin & Bateson 1993).

A linear regression was used to establish whether male body weight could predict dominance status. Data Reduction using a Principal Components Analysis (PCA) was applied to determine which morphological variables best represent body size and explain the variance in the data for further analyses. To correct for size differences among males within each population, the measurements were weighted relative to the number of males in each population. A Generalised Linear Model (GLM) was used to model the relationship between the response variable (number of offspring sired) and the potential explanatory variables (morphology, testosterone concentrations and dominance rank) on male reproductive success. The GLM used a Poisson regression and a log-linked function, which modelled the number of offspring sired (count data) using a Poisson distribution. The raw testosterone measures were not normally distributed and were log-transformed prior to analysis. All statistical analyses were carried out using SPSS 15.0.

Chapter 5 Traits influencing male reproductive success 135

Genetic diversity was estimated for each population by calculating the allelic diversity

(AD), the observed (Ho) and expected (He) heterozygosities using GENALEX 6.0 (Peakall & Smouse 2006). Locus independence and Hardy-Weinberg equilibrium tests were conducted in GenePop 3.4 (Raymond & Rousset 2003) using a Markov chain method (1000 iterations). The statistical significance levels were corrected for multiple comparisons using sequential Bonferroni adjustments (Rice 1989).

To determine the influence of parental relatedness and individual genetic diversity on male reproductive success three methods were used: internal relatedness (IR), a measure that estimates the relatedness of an individual’s parents by weighting the extent of microsatellite allele sharing by the frequency of the alleles present (Amos et al. 2001a); standardised heterozygosity (SH), a measure of individual heterozygosity (Coltman et al. 1999c); and standardised mean d2, a measure of the difference in the number of repeats between two microsatellite alleles at a locus that reflects evolutionary relationships (Coulson et al. 1998). These measures were calculated using ‘IRmacroN’, an Excel© macro (Microsoft Office 2000) written in Visual basic by W. Amos. To determine whether these measures differed from random, the average of each measure was estimated by (i) randomising the genotypes of all males, and (ii) randomising the genotypes of those males that sired offspring. These values were compared with the observed values using a Monte Carlo permutation analysis (10 000 iterations) using PopTools 3.0.3 (Hood 2008). The significance of this relationship was estimated using a linear regression model using randomisation tests appropriate for non-normal and/or non-independent data (Ludbrook & Dudley 1998; Manly 2001). New gradients were estimated from regressions based on the random allocations of male reproductive success to the predictor variables (IR, SH and mean d2) (Manly 2001). The real and randomised t-values were analysed using a Monte Carlo random permutation analysis in PopTools 3.0.3 using 10 000 iterations. The one-tailed probabilities (p-value) of obtaining gradients of the observed magnitude by chance were derived from these data.

Chapter 5 Traits influencing male reproductive success 136

5.3 Results

5.3.1 Male dominance

The dominance rank for each male and the hierarchy for each population are summarised in Table 5-2. The dominance hierarchies were linear in each population (h > 0.6). The hierarchy was significantly linear in population B (h = 0.95), but less so in populations A and C (h = 0.65 – 0.75). The size range and means (± se) for the morphological measurements are summarised in Table 5-3. The size range includes all males (n = 21) and the means for two groups: alpha males (n = 3) and lower ranked males (n = 18). The alpha male in each population was significantly heavier than lower 2 2 ranked males (Population A r = 0.988, F1, 4 = 318.183, p < 0.001; Population B r = 2 0.740, F1, 7 = 19.965, p = 0.003; Population C r = 0.719, F1, 4 = 10.233 0.001, p = 0.033; Figure. 5-1).

Chapter 5 Traits influencing male reproductive success 137

Table 5-2 Summary of the dominance rank (I) for each male, and hierarchy linearity (h) for three semi-free ranging captive populations of eastern grey kangaroos (Macropus giganteus).

Population Male ID Rank* I h A Purple 1 0.906 0.714 Red-yellow 2 0.774 No tag 3 0.667 Mr W 4 0.500 Green-pink 5 0.000 Orange 6 0.000 B Green 1 1.000 0.950 Blue Snr 2 0.764 Orange 3 0.700 Yellow 4 0.683 Red 5 0.522 Purple 6 0.314 Blue Jnr 7 0.133 Pink 8 0.038 White 9 0.027 C Green 1 1.000 0.686 Yellow 2 0.476 Orange 3 0.475 Purple 4 0.458 Red 5 0.324 Blue 6 0.103 *Dominance rank (highest rank = 1 to the lowest rank = 9).

Chapter 5 Traits influencing male reproductive success 138

40 Weight (kg) Weight 20

0 0123456789 Male dominance rank

Figure 5-1 Relationship between male dominance rank and body weight (kg) in eastern grey kangaroos (Macropus giganteus) in three semi-free ranging captive populations (Population A () n = 6; Population B () n = 9; and Population C () n = 6). Dominance rank (highest rank = 1 to the lowest rank = 9).

Table 5-3 Summary of the size range of the 10 morphological traits measured in male eastern grey kangaroos (Macropus giganteus) across all individuals and the mean (± se) for alpha and lower ranked males.

Morphological trait Range Alpha males Lower ranked males Sample size (n) 21 3 18 Body weight (kg) 24.8 - 70.5 55.2 ± 7.9 42.5 ± 3.1 Head length (cm) 20.3 – 25.1 23.6 ± 1.0 23.2 ± 0.3 Forearm length (cm) 19.0 - 35.4 32.5 ± 1.8 27.1 ± 1.2 Forearm circumference (cm) 11.5 - 23.3 20.2 ± 1.6 16.3 ± 0.9 Bicep circumference (cm) 15.0 - 35.3 28.2 ± 4.0 21.4 ± 1.2 Leg length (cm) 46.8 - 65.0 60.9 ± 1.5 56.9 ± 1.3 Pes length (cm) 31.1 - 38.5 35.8 ± 1.5 34.4 ± 0.4 Tail length (cm) 74.7 - 103.5 96.2 ± 3.6 88.3 ± 2.1 Tail circumference (cm) 23.1 - 39.2 34.2 ± 2.8 30.1 ± 1.1 Testis volume (cm3) 9.0 - 42.2 25.5 ± 4.8 24.9 ± 2.2

Chapter 5 Traits influencing male reproductive success 139

5.3.2 Characterisation of microsatellite loci and paternity assignment

All loci were polymorphic and in Hardy-Weinberg equilibrium in each population and there was no evidence for the presence of null alleles. The mean number of alleles ranged from 8 to 12. The average heterozygosity estimates varied between 0.683 and 0.930 (Table 5-4). Allele frequency data for each eastern grey kangaroo population are presented in Appendix 9. Since maternity was known, paternally derived alleles were readily identified. Paternity for 100% of offspring (n = 40) was successfully determined by exclusion since all potential sires were sampled and genotyped.

Table 5-4 Summary of the mean allelic diversity (AD), range of alleles, mean and range of observed (Ho) and expected (He) heterozygosity for the three populations of eastern grey kangaroos (Macropus giganteus) used in this study.

Pop Mean AD Range AD Mean Ho Range Ho Mean He Range He A 12.3 7 - 22 0.771 0.571 – 0.946 0.850 0.779 – 0.930 B 8.1 6 - 12 0.720 0.550 – 0.900 0.771 0.683 – 0.833 C 9.6 6 - 14 0.768 0.526 – 0.895 0.813 0.699 – 0.903

5.3.3 Traits that influence male reproductive success

The PCA of the relative morphological traits (listed in Table 5-3) indicated that all traits except pes length were highly correlated with body weight and therefore a good indicator of body size. Pes length was removed from the model and the new PC1 scores were used in the GLM. The GLM indicated that males with higher reproductive success were significantly larger (p = 0.020), had significantly higher testosterone concentrations (p = 0.021) and held a position of dominance (p = 0.015). It should be noted that the same pattern of testosterone was not present in all populations (Figure 5-2). Alpha males sired significantly more offspring than lower ranked males (Figure 5-3).

Chapter 5 Traits influencing male reproductive success 140

(a) 35 30 25 20 15 10

Testosterone (nmol/L) Testosterone 5 0 12345 (b) 35 30 25 20 15 10

Testosterone (nmol/L) Testosterone 5 0 123456789

Testosterone (nmol/L) Testosterone 5 0 123456 Dominance Rank

Figure 5-2 Relationship between male dominance rank and serum testosterone concentration (nmol/L) in three semi-free ranging captive populations of eastern grey kangaroos (Macropus giganteus) (a) Population A, (b) Population B, and (c) Population C. Dominance rank (highest rank = 1 to the lowest rank = 9).

Chapter 5 Traits influencing male reproductive success 141

100

Offspring sired (%) sired Offspring 20

0 123456789 Male Dominance Rank

Figure 5-3 Percent (%) offspring sired by each male eastern grey kangaroo (Macropus giganteus) of varying dominance rank in three semi-free ranging captive populations (Population A (black) n = 6; Population B (shaded) n = 9; Population C (white) n = 6). Dominance rank (highest rank = 1 to the lowest rank = 9).

5.3.4 Relatedness, heterozygosity and reproductive success

There was a significant negative relationship between parental relatedness and reproductive success (Figure. 5-4). Sires were less related to dams than non-sires (IR: p = 0.029). Sires were significantly more heterozygous than non-sires (SH: p = 0.018). However there was no significant difference in heterozygosity between sires and non- sires using mean d2 values (p = 0.762).

Chapter 5 Traits influencing male reproductive success 142

* 1.2 1 0.8

0.6 0.4 * Mean value 0.2

0 IR SH Mean d2 -0.2 Relatedness parameter

Figure 5-4 The mean (± se) difference in individual heterozygosity between sires () and non-sires () for internal relatedness (IR), standardised heterozygosity (SH) and mean d2 in the eastern grey kangaroo (Macropus giganteus). * indicates a significant difference (p < 0.05).

5.4 Discussion

This is the first study to demonstrate with molecular data that alpha male kangaroos sire the most offspring and to identify traits influencing male reproductive success. The probability of acquiring a position of dominance was significantly influenced by male body weight and body size, in relation to other males present. Dominant males sired significantly more offspring than smaller, lower ranked males, and had significantly higher testosterone concentrations. Sires were significantly more heterozygous and genetically dissimilar to females, than non-sires. These results show that there are several interrelated factors that contribute to male reproductive success in the eastern grey kangaroo.

5.4.1 Male dominance

The eastern grey kangaroo mating system is based on males contesting for access to females which is consistent with their pronounced sexual size dimorphism (Andersson 1994). Male kangaroos compete intensely for dominance positions and the probability

Chapter 5 Traits influencing male reproductive success 143 of attaining dominance increased significantly with larger body weight and body size. Limited breeding opportunities for males within a mob coupled with the high reproductive potential of the alpha male should result in strong selective pressure for males to increase their body weight relative to competitors, as well as control the number of prospective competitors within the mob. As predicted, the alpha male in each population was the heaviest and the largest. A clear dominance hierarchy was observed in all three populations, although it was only significantly linear in Population B (h = 0.95). These data suggest that alpha males can outcompete and exclude smaller males as has been found in other polygynous species (Clutton-Brock et al. 1988; McElligott et al. 2001). However, the importance of fighting ability could not be directly addressed in this study as too few incidents of ritualised fighting occurred. Observations indicate that when lower ranked males in a mob are ready to contest for the alpha position, they increasingly challenge the alpha male. These contests continue until one party concedes defeat, and can be fatal (pers. obs.). During this study the second ranked males in Populations A and C were increasingly challenging the alpha male for his position and subsequently both displaced and replaced the alpha male from his position during the breeding season following the cessation of behavioural observations for this study (E. J. M. unpublished data).

5.4.2 Male reproductive success, dominance, body size and testosterone

The pattern of reproductive success revealed from the genetic data was consistent with previous behavioural studies for the eastern grey kangaroo (see Introduction). Overall, alpha males had significantly higher reproductive success than lower ranked males, and on average sired 54% of all offspring. This level of reproductive success was comparable to that reported in other polygynous species. For example, alpha male tammar wallabies (Hynes et al. 2005; Miller et al. in press), as well as northern (Mirounga angustirostris) and southern elephant seals (M. leonina) (Hoelzel et al. 1999) sired around 50% of offspring. In other species with a multi-male mating system the alpha male’s reproductive success varied between 13% and 93% (Clutton-Brock & Isvaran 2006). These results provide empirical data to support the priority-of-access model, and conform to the prediction that selection favours phenotypic adaptations in males to enhance their ability to succeed in male-male contest to gain access to females.

Chapter 5 Traits influencing male reproductive success 144

The strength of the association between dominance rank and reproductive success depends on various factors, including the ability of males to monopolise access to receptive females, the relative success of alternative male mating strategies and the impact of female mate choice (Bercovitch 1991). Although the alpha males sired the majority of offspring, complete monopolisation of paternity is difficult and some lower ranked males achieved successful copulations when multiple females were in oestrus. This was most obvious in Population C where the second rank male was verging on toppling the alpha male from his position of dominance towards the end of the breeding season and achieved considerable reproductive success. The alpha male in Population C also developed a health problem (degenerative hip condition) that potentially impaired his ability to mate with females, but nevertheless distinctly maintained behavioural dominance within the mob.

Male eastern grey kangaroos that sired offspring had significantly higher testosterone concentrations than non-sires. However, the patterns of testosterone concentration across males of varying rank in Population B showed a more erratic pattern to that observed in Populations A and C. The outcome of agonistic exchanges may be the product of several other interacting factors. The production of a range of hormones can also be affected by antagonistic encounters and social subordination (Clark & Faulkes 1998). A number of studies have shown that testosterone varies seasonally long-term) and with social stimuli (short-term) (Wingfield et al. 1990). Thus it is likely that testosterone concentration varies temporally within individuals, and populations. Also, testosterone is secreted in a pulsatile pattern in response to the pulsatile nature of luteinising hormone secretion, therefore one off blood samples may happen to coincide with a peak or trough in testosterone concentrations. Due to such variation, testosterone levels may not always be indicative of mating strategies or reproductive success.

5.4.3 Male reproductive success and genetic traits

This study found that females successfully mated with males that were more genetically dissimilar to them than expected by chance, and that sires were significantly more heterozygous than non-sires. Genetic dissimilarity may be selected for via cryptic female mate choice, and there is evidence that females prefer

Chapter 5 Traits influencing male reproductive success 145 genetically dissimilar mates (Amos et al. 2001a; Parrott et al. 2007; Parrott et al. 2006). This preference may be a mechanism to avoid genetic incompatibility and inbreeding which can lead to decreased offspring fitness (Tregenza & Wedell 2000). Inbreeding avoidance has also been linked to sex-biased dispersal (Dobson & Jones 1985; Moore & Ali 1984; Pusey 1987) and the dispersal of males in the EGK may reduce the chance of inbreeding in wild populations, however females generally do not disperse. Female preference for genetically dissimilar males may be a mechanism for inbreeding avoidance within a mob as alpha males will sire many offspring during their reign which can span several breeding seasons (Bolitho et al. unpublished data).

There was a significant relationship between male reproductive success and standardised heterozygosity, but not mean d2. Mean d2 has been shown to be informative under limited circumstances (Hedrick et al. 2001a) and the results from this study are consistent with other findings (Amos et al. 2001a). Therefore mean d2 will be excluded from further discussion. The significant relationship between male reproductive success and standardised heterozygosity suggests individual fitness increases with higher levels of genetic diversity. More heterozygous males may be more likely to become an alpha male if they are fitter, thus achieving higher reproductive success. Research in other taxa such as marine mammals (Amos et al. 2001b) and birds (Hansson et al. 2004) has shown females prefer genetically diverse mates for siring their offspring. Females may choose the most heterozygous and genetically dissimilar males to sire offspring to pass on their ‘good’ genes. According to the ‘good genes’ hypothesis males with exaggerated traits must have good genes (Hamilton & Zuk 1982) to invest in such traits, despite the handicap of investing in that trait. Females who choose such males are thought to provide their offspring with the advantage of inheriting the good genes. The advent of molecular biology has enabled us to examine this model on a genetic level. Research has shown that higher levels of heterozygosity are not only an indicator of male quality (Seddon et al. 2004), but also individuals with higher heterozygosity have been linked to parasite resistance (Acevedo-Whitehouse et al. 2006), increased survivorship (Bean et al. 2004) and increased fecundity (Ferreira & Amos 2006). However, it should be noted that this study was designed to examine traits influencing male reproductive success and not female mate choice directly. It has been suggested that individuals not only assess a

Chapter 5 Traits influencing male reproductive success 146 potential mates quality (immuno-competence) but also how well they complement their own genes (Piertney & Oliver 2006).

The major histo-compatibility complex (MHC) is an integral component of the vertebrate immune system (Knapp 2005) and evidence suggests that it plays a role in mate choice (Piertney & Oliver 2006). Studies have shown that individuals mate disassortatively with respect to MHC genotype (Freeman-Gallant et al. 2003; Olsson et al. 2003). MHC-based mate choice would facilitate inbreeding avoidance, as well as increase offspring fitness by maximising immunological potential (Piertney & Oliver 2006). Empirical data from pheasants has shown that male spur length is positively correlated with age, body size and viability. Females prefer to mate with males with longer spurs, and by doing so, they improve chick survival rate (von Schantz et al. 1989). Subsequent studies have shown MHC haplotypes for both class I and II correlate with spur length and male viability. This indicates that females discriminate among males based on ‘good genes’ (von Schantz et al. 1997; von Schantz et al. 1996). Research examining the relationship between female mate choice and the MHC in eastern grey kangaroo are in progress.

5.4.4 Implications for management

The eastern grey kangaroo is overabundant in many parts of its range (DEH 2007) and overabundant populations can be managed through commercial and non-commercial harvesting, translocation, restricting dispersal though fencing, reduced fertility through vasectomy, fertility control or sterilisation, and euthanasia (Caughley 1987; Coulson 1998; Herbert 2007). Consistent with other research, body size in the EGK has an important role in their ecology and evolution. As body size correlates with many aspects of an individual’s biology, it is considered an important trait (Fenberg & Roy 2007). Therefore biological and genetic principles should be considered when devising management plans. Kangaroo culling operations often selectively harvest large males (DEH 2007). It has been shown in numerous species that selective harvesting of males can have effects on populations, such as selecting for smaller males (Allendorf et al. 2008; Coltman et al. 2003), reproductive collapse (Fairall 1985; Milner-Gulland et al. 2003; Milner-Gulland & Mace 1991; Mysterud et al. 2002) and changes in population structure (Allendorf et al. 2008; Nussey et al. 2006). Kangaroo managers are aware of

Chapter 5 Traits influencing male reproductive success 147 these potential consequences and have attempted to model the outcomes, without access to important data, such as male reproductive success. The results from this study will enable managers to construct more informative models. In other areas of the kangaroos range, where culling is not a management option for sociopolitical or practical reasons, fertility control has been proposed as a potential management strategy (Adderton Herbert 2004). The idea of selectively targeting the dominant males via vasectomy, whereby they continue to secrete testosterone and hence maintain their dominance status, has been proposed as a potential cost effective strategy. But, given that females are seasonally polyoestrous (Tyndale-Biscoe & Renfree 1987), and smaller males can sire offspring, this is unlikely to be an effective strategy.

5.4.5 Conclusions

This is the first study to provide empirical data to support the long-held assumption that the alpha male sires the most offspring in the EGK. This research identified several interrelated traits influencing male reproductive success, including dominance status, body weight, body size and testosterone concentrations. Males that enjoyed greater reproductive success were also more heterozygous and the most genetically dissimilar to the dams they sired offspring with. The extent to which female mate choice exerts control over male reproductive success is unclear but warrants further investigation. As male body size is an important aspect of EGK ecology and evolution, management and conservation strategies should consider this when formulating plans to control overabundant populations. These findings provide important information on reproductive strategies in kangaroos, as well as further contributing to our understanding of male reproductive success in marsupials.

Chapter 6 Validation of body condition indices 148

Chapter 6 Validation of body condition indices using serum biochemistry and haematology in a Macropodid species

6.1 Introduction

Body condition refers to the energetic state of an individual. Individuals in ‘good condition’ are thought to have greater energy reserves than those in ‘poor condition’ (Millar & Hickling 1990). In evolution and ecology, body condition has been identified as an important determinant of an individual’s fitness as it encompasses several attributes including an individual’s immune response (Gleeson et al. 2005), ability to defend itself a territory, foraging ability (Humphreys et al. 1984), migratory behaviour (Tsvey et al. 2007) and reproductive success (Bercovitch & Nürnberg 1996; Guinet et al. 1998; Hodges et al. 1999). The effect of body condition differs between the sexes. For example, in female mammals, a higher body condition has been associated with higher reproductive fitness traits such as litter mass, number of litters, and lifetime reproductive success (Atkinson & Ramsay 1995; Dobson & Michener 1995). During the breeding season when males are competing for access to females, higher fat reserves can be beneficial due to the amount of energy they expend during male-male competition (Deutsch et al. 1990). Body condition indices (BCI) have also been used when considering population management and individual animal welfare. For example, knowledge of an individual’s health can be useful for predicting survival in reintroductions and translocations (Mathews et al. 2006). In accordance with the Prevention of Cruelty to Animals Act 1985, individuals or populations deemed to be in poor condition and suffering, are often culled. Given the broad use of condition indices

Chapter 6 Validation of body condition indices 149 in both evolutionary ecology and management, validation of such indices should be paramount. This study aims to validate five commonly used BCIs using serum biochemistry and haematology in an overabundant Australian marsupial species whose body condition is relevant to management decisions.

There are numerous methods for calculating body condition. Measuring fat reserves directly is lethal, for example the collection of kidney fat and/or bone marrow (Shepherd 1987). For many scientists and wildlife managers lethal sampling is rarely desirable and consequently a range of non-lethal indices have been developed to measure body condition using morphological parameters that are thought to correlate with an individual’s nutritional state or fat reserves (Weatherhead & Brown 1996). There are several non-destructive methods of calculating a BCI and most fall within two chief categories: ratio and residual indices. Ratio indices are generally calculated as body mass divided by a linear measurement of body size and have been used in mammals (e.g., Bercovitch et al. 2003), birds (e.g., Whitfield et al. 1999) and reptiles (e.g., Dickinson & Fa 2000). There are several variants of this method such as raising the linear measurement to a statistical power. The most common is the Quetelet’s index (also referred to as body mass index) that divides mass by a squared linear measurement (see Eknoyon 2008 for review). Another popular method is Fulton’s condition index (K) (Ricker 1975) whereby the linear measurement is raised to the power of three as this assumes the mass and linear dimensions increase isometrically (Cone 1989). This method was originally developed for fisheries science (e.g., Neilson et al. 1986), although it has been applied in mammals (e.g., Bakker & Main 1980) and reptiles (Vervust et al. 2008). Generally for ratio indices, a higher ratio value is thought to reflect a better body condition than a lower ratio (Schulte-Hostedde et al. 2001a). However, there is evidence that ratio indices lack independence with respect to body mass and body size, which can compromise results (Jakob et al. 1996).

Residual indices differ from ratio indices in that the body mass is regressed on a measure of body size. Residual indices are considered a more reliable index of body condition than ratio indices because they do not vary with size (Green 2001; Jakob et al. 1996). Currently the most frequently used BCI is an ordinary least squares regression (OLS), whereby body mass is regressed on a linear measure indicative of

Chapter 6 Validation of body condition indices 150 body size (skeletal measurement) (Cone 1989; Green 2001; Jakob et al. 1996). Individuals with a positive residual are thought to be in a better condition than predicted for their body size, and those with a negative residual value are considered to be in comparatively poorer condition. However, the method of analysis for residual condition indices has been much debated during the past two decades (Green 2001; Jakob et al. 1996; Kotiaho 1999; Schulte-Hostedde et al. 2005b). Jakob et al. (1996) recommended OLS residuals be used as a condition index, but were later criticised as many studies that use this method violate a number of assumptions (Green 2001; Kotiaho 1999). Green (2001) suggested the use of a model II regression may be a suitable alternative to OLS as this regression model reduces the major axis of the regression (RMA), to account for larger individuals that can increase their body condition at a proportionally higher rate than smaller individuals. Recent analyses by Schulte-Hostedde et al. (2005b) suggest that OLS residuals are in fact better estimates of body condition than RMA residuals.

Alternative methods to measure body condition include the measurement of blood parameters such as serum biochemistry and haematology. Blood chemistry can be used to assess the health and well-being of animals. Although some parameters such as hormone concentrations can fluctuate in minutes, other parameters such as white cell counts are more constant and can reflect constant aspects of an individual’s condition, for example parasite load or infectious status. Many studies have used these parameters to examine how they change with season, age and/or sex (McKenzie et al. 2002; Viggers & Lindemayer 2001). Some studies have utilised blood parameters in conjunction with BCIs to assess an animal’s health and condition (Beldomenico et al. 2008; Sanchez-Guzman et al. 2004), but they have not applied these data to determine whether BCIs accurately reflect these other health parameters.

The eastern grey kangaroo (Macropus giganteus) is a gregarious Australian macropodid marsupial that lives in large mixed sex groups and have an extensive and almost continuous distribution throughout eastern Australia (Poole 2002). They are an overabundant species, considered desirable in many parklands and reserves, but are often not tolerated outside of protected areas (Garrott et al. 1993). Currently eastern grey kangaroos are managed through several techniques including euthanasia,

Chapter 6 Validation of body condition indices 151 translocation, fertility control, commercial and non-commercial harvesting and exclusion fencing of areas. In areas where they are overabundant, the eastern grey kangaroo population may outstrip the available resources, leading to a loss of condition and mortality in some animals. This is particularly a problem with populations that have been deliberately or inadvertently confined to areas through fence construction and/or urban development (Adderton Herbert 2004). Overabundant or expanding populations also impact other species negatively (Garrott et al. 1993). Monopolisation of resources by overabundant species can reduce both plant and animal diversity, changing the species composition or relative abundance of sympatric species, introducing and/or spreading infectious diseases and parasites, and in some cases leading to local extinctions (Coulson 1998; Garrott et al. 1993; Noss 1990; Temple 1990). The formal policy framework for kangaroo management in Australia stipulates that culling should take place to alleviate or prevent the suffering of individuals (DEH 2007). The health and condition of the population is monitored, both prior to the cull to assess its necessity, and after the cull to monitor its success (DEH 2007). In kangaroos, two commonly used methods of assessing body condition are visual assessment and measuring the circumference of the base of the tail where fat is deposited, but the latter has poor repeatability and accuracy (Bakker & Main 1980; Moss & Croft 1999). Given the highly politicised nature of kangaroo culling along the east coast of Australia, it is important that the measures used to justify kangaroo culling on welfare grounds are appropriately validated. The aim of this investigation was to compare and validate five common methods (Table 6-1) of calculating body condition using serum biochemistry and haematology parameters in the eastern grey kangaroo. A secondary aim was to provide a reference range for serum biochemistry and haematology parameters for adult eastern grey kangaroos in ‘good’ and ‘poor’ condition.

6.2 Materials and Methods

6.2.1 Study populations

The two contrasting eastern grey kangaroo populations selected for this study were broadly classed as ‘good’ condition (Population A) and ‘poor’ condition (Population B). Population A was located on the outskirts of Sydney at the University of New South Wales Research Facility, New South Wales (NSW), Australia (33º 55’S, 150º

Chapter 6 Validation of body condition indices 152

57’E). The kangaroos were maintained in nine hectares (ha) of enclosed natural bushland that was two thirds open grassland and one third dry sclerophyll forest. Within the enclosure there was a supplementary feed bin and several water troughs. Animals were fed on a commercial grain diet of ‘kangaroo pellets’ ad libitum (YF Feeds, Young, NSW) and supplemented with lucerne three times a week. The population was treated for intestinal parasites every two months with Equiban granules (Pfizer, Australia). Population B was a wild population located in a 156ha reserve on the outskirts of Melbourne, Victoria, Australia (38º 00’S, 144º 24’E) which consists of a variety of habitats, including wetlands, grasslands and farmland. Kangaroos in the reserve were at a density of 4.5 kangaroos per ha (M. Wilson, unpublished data), and the population received no supplementary feeding or preventative medical treatment. The population was considered to be in poor condition from overgrazing and was the subject of a culling operation.

6.2.2 Sample collection

Adults were used for this study as growth in juveniles may confound the relationship between body size and body mass. For females, the pouch was checked to confirm that the individuals were sexually mature (i.e. they had everted teats or current evidence of reproductive activity) and their reproductive status was classified as either: i) sexually mature with no young present; ii) suckling a young confined to the pouch (pouch young, PY); iii) suckling a young-at-foot (YAF), as evidenced by an elongated teat and active mammary gland; or iv) having recently weaned a young, as indicated by a regressing teat and inactive mammary gland. It has been well documented that body condition varies significantly with season (Beldomenico et al. 2008; Moss & Croft 1999; Stirrat 2003; Viggers et al. 1998) so in order to exaggerate the difference between these populations, Population A were sampled during late spring (November) when they should be in better condition, and Population B were sampled during late autumn (May).

In Population A, adult males (n = 11) and large females (n = 16) were chemically restrained using a blow pipe and dart (1CC, ¾”, volume 1ml, PA 17703, Pneu-Dart Inc., Williamsport) containing tiletamine HCl in combination with zolazepam HCl (Zoletil 100®, Virbac (Australia) Pty. Ltd., Peakhurst, NSW), at a dose rate of 5mg/kg

Chapter 6 Validation of body condition indices 153 delivered into the thigh muscle of the kangaroo at a distance of up to 20m. Following drug administration animals were initially monitored from a distance until the drug took effect. When an animal was sufficiently sedated (usually after 5 – 10 min) it was placed in a jute sack in the shade and immediately processed. Recovery following anaesthesia usually took one hour. Animals were placed in a quiet, shady location with their eyes covered until fully recovered. The kangaroos in Population B (males n = 9; females n = 15) were shot and taken to a mobile field laboratory immediately where the kangaroos were sexed, weighed, and samples collected.

For both populations, each animal was weighed using a portable 100kg Salter suspended dial scale (SA-235/9, Wedderburn, accuracy 0.05kg). Two morphological measurements were collected from the left leg of the animal with the pes (foot) at 90° to the tibia and the tibia at 90° to the femur. Pes length was measured from the base of the heel to the tip of the main toe (fourth digit), excluding the toenail. Leg length was measured from the most distal point of the femur to the base of the heel.

Blood samples were collected from the lateral tail vein (Plate 6-1). A small area of the basal lateral tail was clipped and swabbed with 70% ethanol. Blood samples (7ml) were collected using a 21 gauge winged infusion set (Surflo, Terumo Corporation, Japan), a 10ml syringe and a rubber tubing tourniquet. Following withdrawal of the needle, light digital pressure was applied to the site to assist haemostasis. Three ml of blood was transferred immediately into serum separation tubes (Vacuette, Greiner Labortechnik, Austria) and inverted gently for biochemical analysis. Four ml of blood was transferred into a tube containing ethylene diamine tetraacetic acid (EDTA; Vacuette, Greiner Labortechnik, Austria), rolled gently to mix the blood and EDTA, and placed on ice to prevent clotting. Samples were collected and processed within 24 hours of sampling by IDEXX Laboratories (Australia). As Population B was shot, some blood samples were collected directly from the chest cavity. Opportunistic sampling of the left kidney was conducted for all individuals in Population B to calculate KFI to further validate the condition of the population.

Chapter 6 Validation of body condition indices 154

6.2.3 Biochemical and haematological analyses

Diagnostic testing was conducted by IDEXX Laboratories (Australia) using an Olympus AU400 for a complete serum biochemistry profile and LaserCyte® Haematology Analyser for a compete haematology profile. Due to the large number of variables in each profile, a subset of variables that was considered meaningful in macropods was selected to examine the health of eastern grey kangaroos (in consultation with a veterinarian, Dr Robert Johnson, at Taronga Zoo, Sydney, Australia). Table 6-2 lists the selected blood parameters and how they fluctuate with respect to some common ailments, such as poor nutritional state and infection.

Table 6-1 Summary of the body condition indices used to compare two populations of eastern grey kangaroos (Macropus giganteus) representing a population in ‘good’ condition and a population in ‘poor’ condition.

Body Condition Index Formula Ratio index Mass/length Quetelet’s index Mass/(length)2 Fulton’s condition factor (K) Mass/(length)3 Ratio.2 Index (mass)2/length Ordinary Least Squares regression (OLS) Mass vs length

Chapter 6 Validation of body condition indices 155

6.2.4 Kidney fat index

A kidney fat index (KFI) was calculated for Population B. This method was first described by Riney (1955), and adapted for kangaroos by Caughley (1962), as a measure of physical condition. The KFI is the relative weight of the fat tissue around the kidney to the kidney weight. The fat tissue was trimmed off the kidney, and the relative weight of the fat to the kidney was determined by dividing the perirenal fat by the kidney weight and multiplying by 100.

Table 6-2 Summary of the main causes of fluctuation in the selected serum biochemistry and haematology parameters used in this study. The arrows denote an increase () or decrease () in the circulating levels of analytes (Meyer & Harvey 1998) . state Anemia Chronic Chronic Chronic Infection Diarrhoea dysfunction dysfunction haemorrhage inflammation Liver disease/ disease/ Liver Poor nutritional nutritional Poor damage Muscle Blood parameter disease/ Kidney Serum biochemistry Urea (mmol/L) Creatinine (mmol/L) Total Protein (g/L) Albumin (g/L) Globulin (g/L) ALP(IU/L) AST (IU/L) ALT (IU/L) CK (IU/L)

Haematology Red blood cells (x1012/L) Haemoglobin (g/L) Haematocrit (L/L) 9 White cell count (x10 /L) ALP = alkaline phosphatase; AST = aspartate aminotransferase; ALT = alanine aminotransferase; CK = creatine kinase.

Chapter 6 Validation of body condition indices 156

6.2.5 Data Analysis

Prior to analysis, a two independent samples t-test using the Mann-Whitney test statistic was used to detect whether there was a significant difference in the biochemistry, haematology or body condition between lactating and non-lactating females within each population. To test the hypothesis that Population A was in better health than Population B, the health of each population was inferred from the nine serum biochemistry and four haematology parameters using an independent-samples t- test to examine whether there were sex-specific differences between sexes within each population, and between each population. The Levene test statistic tests the assumption of equal variances. The appropriate test statistic was used for each variable depending on whether the data met this assumption or not. Females and males were analysed separately for all further analyses.

Two populations representing both ends of the condition spectrum were deliberately selected for this study to allow the validation and comparison of five BCIs across populations. To compare the relationship between populations for each sex, regression equations were derived from the sex-specific pooled population data (Krebs & Singleton 1993). The body mass (BM) and length (leg and pes) variables for all individuals were transformed to natural logarithms. The relationships between the ln- transformed variables were described for each sex using an ordinary least squares (OLS) regression. The relationship between BM and length was described as ln BM =

1 * ln (length) + 0, with 1 representing the intercept and 0 representing the slope of the line. Before the mass and length residuals could be used for calculating the BCIs, they had to meet three main assumptions (Green 2001). First, a linear regression was used to verify the assumption that there was a linear relationship between body mass and length. Second, to determine whether leg and pes length were reliable indicators of body size, a linear regression was used to test the relationship between leg length and skeletal (bone) mass in the grey kangaroo from museum specimens. Last, to ensure that the residuals that represent body condition were independent of body size, a linear regression was used to examine whether larger animals had higher BCI values. To test for any differences in mean skeletal size between populations, sex-specific paired t- tests were conducted.

Chapter 6 Validation of body condition indices 157

Five BCIs (Table 6-1) were calculated for each individual. To calculate the OLS residuals, the ln-transformed body weight (kg) was regressed against the ln- transformed leg length (cm). The resulting equation was used to predict the body weight for each individual. The residuals used were the difference between the observed and predicted body weights for each individual (Krebs & Singleton 1993). Generally, individuals with a positive residual are considered to be in better condition than an individual with a negative residual (Schulte-Hostedde et al. 2005b). An independent-sample t-test was used to examine whether there were significant sex- specific differences in body condition between the two populations.

To examine whether the BCIs calculated were an accurate reflection of an individual’s health based on their blood chemistry, a generalised linear model (GLM) was used. The GLM procedure allows for the dependant variable to have a non-normal distribution, and tests for the main effects contributing to the model. The dependent variables (each BCI) were analysed separately for the biochemistry and haematology. When there are too many variables, GLM models can have difficulty interpreting the data, therefore a subset of biochemistry parameters were selected based on their biological meaningfulness (Table 6-2). Five of the nine biochemistry parameters were analysed separately from the four haematology parameters. The model used a Poisson distribution with a log link function, and Wald chi-squared test statistic. All statistical analyses were conducted in SPSS 15.0.

6.3 Results

The independent samples t-test using the Mann-Whitney test statistic showed no significant difference between lactating and non-lactating females for any blood chemistry parameter or any BCI within each population (p > 0.05). Therefore the non- lactating and lactating females were pooled in subsequent analyses. There was a significant difference between males and females in Population A for all blood parameters (p < 0.05) except total protein, albumin, globulin and white cell count (p > 0.05). In contrast, only total protein and globulin were significantly different (p < 0.05) between females and males in Population B. Given these variable differences between sexes within the different populations, males and females were analysed separately.

Chapter 6 Validation of body condition indices 158

6.3.1 Serum biochemistry and haematology

The independent samples t-test supported the hypothesis that Population A was generally in better health than Population B based on their blood parameters (Tables 6-2 and 6-3(a) females and 6-3(b) males). In comparison to females in Population A, females in Population B had significantly lower levels of creatinine (t = 5.506, p = 0.000), total protein (t = 3.096, p = 0.006) and albumin (t = 3.700, p = 0.001); significantly higher levels of AST (t = -5.514, p = 0.000), ALT (t = -2.513, p = 0.022) and CK (t = -6.640, p = 0.000); and significantly lower red blood cells (t = 6.720, p = 0.000), haemoglobin (t = 2.948, p = 0.007), haematocrit (t = 6.075, p = 0.000) and white blood cells (t = 4.396, p = 0.000). A similar pattern of differences was observed in the males. Males in Population B had significantly lower total protein (t = 6.030, p = 0.000), albumin (t = 3.564, p = 0.002) and globulin (t = 6.986, p = 0.000) and significantly higher AST (t = -3.016, p = 0.017) and CK (t = -2.964, p = 0.018). Males in Population B also had significantly lower red blood cells (t = 6.039, p = 0.000), haemoglobin (t = 5.534, p = 0.007), haematocrit (t = 5.122, p = 0.000) and white blood cells (t = 4.591, p = 0.000).

Chapter 6 Validation of body condition indices Chapter 6 Validation of body condition indices 159 Table 6-3 Summary of the mean (± se) and reference range of the serum biochemistry and haematology parameters for two eastern grey kangaroo (Macropus giganteus) populations. Population A = ‘good’ condition, and Population B = ‘poor’ condition in (a) females, and (b) males. ‘Pooled’ represents the mean calculated from pooled population data from A and B and therefore depicts the mean values across the full spectrum of body conditions in this species.

(a) ‘Good' condition (n = 16) Pooled ‘Poor' condition (n = 15) Parameter Range Mean ± se Mean ± se Range Mean ± se Serum biochemistry Urea (mmol/L)* 6.8 -14.7 10.79 ± 0.58 12.37 ± 0.47 11.8 - 18.5 14.05 ± 0.45 Creatinine (mmol/L)* 0.09 - 0.19 0.13 ± 0.01 0.11 ± 0.01 0.07 - 0.12 0.09 ± 0.004 Total Protein (g/L)* 58 - 74 65.25 ± 1.03 61.52 ± 1.28 45 - 72 57.53 ± 1.98 Albumin (g/L)* 28 - 40 34.63 ± 0.86 32.19 ± 0.79 22 - 35 29.60 ± 0.98 Globulin (g/L) 24 - 39 30.63 ± 1.23 29.32 ± 0.99 14 - 37 27.93 ± 1.52 ALP (IU/L) 104 - 497 307.63 ± 32.22 315.16 ± 33.24 110 - 975 323.20 ± 60.80 AST (IU/L)* 28 - 92 49.38 ± 4.35 171.03 ± 31.31 80 - 780 300.80 ± 44.61 ALT (IU/L)* 27 - 69 50.69 ± 2.78 57.84 ± 3.74 34 - 131 65.47 ± 6.72 CK (IU/L)* 137 - 899 359.75 ± 48.37 1781.42 ± 341.64 341 - 5936 3297.87 ± 442.69

Haematology Red blood cells (x1012/L)* 4.0 - 6.1 4.82 ± 0.15 3.70 ± 0.24 0.7 - 4.6 2.73 ± 0.26 Haemoglobin (g/L)* 145 - 166 122.73 ± 10.05 100.94 ± 7.10 22 - 137 82.07 ± 7.31 Haematocrit (L/L)* 0.32 - 0.52 0.41 ± 0.01 0.32 ± 0.02 0.06 - 0.42 0.24 ± 0.02 White cell count (x109/L)* 4.4 - 10.4 6.92 ± 0.43 4.57 ± 0.60 0.7 - 13.0 2.53 ± 0.79 * indicates a significant difference between populations (p < 0.05). ALP = alkaline phosphatase; AST = aspartate aminotransferase; ALT = alanine aminotransferase; CK = creatine kinase. Chapter 6 Validation of body condition indices 160

(b) ‘Good' condition (n = 11) Pooled ‘Poor' condition (n = 9) Parameter Range Mean ± se Mean ± se Range Mean ± se Serum biochemistry Urea (mmol/L)* 5.3 - 13.3 8.93 ± 0.63 11.04 ± 0.72 9.9 - 18.6 13.61 ± 0.81 Creatinine (mmol/L) 0.07 - 0.14 0.10 ± 0.007 0.09 ± 0.01 0.05 - 0.13 0.09 ± 0.01 Total Protein (g/L)* 58 - 71 65.82 ± 1.18 57.35 ± 2.63 28 - 63 47.00 ± 3.15 Albumin (g/L)* 27 - 39 33.91 ± 1.05 30.80 ± 1.23 17 - 36 27.00 ± 1.72 Globulin (g/L)* 28 - 39 31.91 ± 0.96 26.55 ± 1.59 11 - 27 20.00 ± 1.48 ALP (IU/L) 204 - 930 556.18 ± 67.51 471.50 ± 53.35 143 - 828 368.00 ± 74.98 AST (IU/L)* 40 - 111 61.45 ± 5.67 188.65 ± 52.04 101 - 936 344.11 ± 93.56 ALT (IU/L) 54 - 92 71.09 ± 3.67 64.95 ± 5.60 13 - 106 57.44 ± 11.49 CK (IU/L)* 193 - 757 502.00 ± 49.65 1427.20 ± 382.69 417 - 6962 2558.00 ± 691.77

Haematology Red blood cells (x1012/L)* 4.6 - 8.7 6.34 ± 0.38 4.83 ± 0.47 1.2 - 5.7 2.98 ± 0.41 Haemoglobin (g/L)* 122 - 215 165.91 ± 8.47 131.05 ± 11.14 34 - 162 88.44 ± 11.53 Haematocrit (L/L)* 0.36 - 0.78 0.53 ± 0.04 0.41 ± 0.04 0.10 - 0.50 0.25 ± 0.04 White cell count (x109/L)* 3.0 - 10.3 7.38 ± 0.69 5.38 ± 0.69 0.8 - 6.9 2.92 ± 0.66 * indicates a significant difference between populations (p < 0.05). ALP = alkaline phosphatase; AST = aspartate aminotransferase; ALT = alanine aminotransferase; CK = creatine kinase

Chapter 6 Validation of body condition indices 161

6.3.2 Validation of assumptions for OLS

Table 6-4 provides a summary of the morphological measurements for females and males in both populations. There was a significantly linear relationship between ln- transformed body weight (kg) and ln-transformed leg length (cm) (p < 0.05; Figure 6- 1(a) females, and 6-1(b) males) and ln-transformed body weight (kg) and ln- transformed pes length (cm) (p < 0.05; Figure 6-2(a) females and 6-2(b) males). There was a significantly linear relationship between body weight and both leg length (p < 0.05), and pes length (p < 0.05), but leg length was a better predictor of body weight and was used subsequently in BCI calculations. The second assumption of OLS could not be tested rigorously due to the small number of adult grey kangaroo skeletons available (n = 4). To satisfy the third assumption for OLS, a linear regression was conducted to assess the relationship between the BCI OLS values and body weight in females and males to test for a size bias. There was no significant relationship between 2 the OLS BCI and body weight for females (F1, 28 = 2.194, r = 0.099, p = 0.091) or 2 males (F1, 18 = 0.868, r = 0.039, p = 0.364).

Table 6-4 Summary of morphological measurements (mean ± se) for females and males in two populations of eastern grey kangaroos (Macropus giganteus) (Population A = ‘good’ condition, and Population B = ‘poor’ condition). * denotes a significant difference between populations (p < 0.05).

Females Males Measure Pop A Pop B Pop A Pop B Body weight (kg) 26.7 ± 0.9* 20.9 ± 1.7* 39.6 ± 4.0 32.0 ± 5.2 Leg length (cm) 48.6 ± 0.5* 44.1 ± 1.4* 54.2 ± 1.7 51.7 ± 3.1 Pes length (cm) 29.7 ± 0.2 29.7 ± 0.4 33.7 ± 0.7 33.3 ± 1.2

Chapter 6 Validation of body condition indices 162

(a)

1.60 y = 3.0929x - 3.7999 2 1.50 r = 0.9069 ) 1.40

1.30

1.20 Log weight (cm weight Log

1.10

1.00 1.54 1.56 1.58 1.60 1.62 1.64 1.66 1.68 1.70 1.72 Log leg length (cm)

(b)

1.90 y = 3.0733x - 3.768 1.80 r2 = 0.9541 1.70

) 1.60

1.50 1.40

Log weight (kg weight Log 1.30 1.20

1.10 1.00 1.60 1.65 1.70 1.75 1.80 1.85 Log leg length (cm)

Figure 6-1 Graphs of the ln-transformed data showing the relationship between body weight (kg) and leg length (cm) for (a) female (n = 31) and (b) male (n = 20) eastern grey kangaroos (Macropus giganteus) from two populations.

Chapter 6 Validation of body condition indices 163

(a)

1.60 y = 4.2997x - 4.9678 2 1.50 r = 0.4654 ) 1.40

1.30

1.20 Log weight (cm weight Log

1.10

1.00 1.40 1.42 1.44 1.46 1.48 1.50 1.52 1.54 Log pes length (cm)

(b)

1.90 y = 4.4077x - 5.1974 1.80 r2 = 0.7482 1.70

) 1.60 1.50

1.40

Log weight (kg weight Log 1.30 1.20 1.10

1.00 1.40 1.45 1.50 1.55 1.60 Log pes length (cm)

Figure 6-2 Graphs of the pooled population ln-transformed data showing the relationship between body weight (kg) and pes length (cm) for (a) female (n = 31) and (b) male (n = 20) eastern grey kangaroos (Macropus giganteus) from two populations.

Chapter 6 Validation of body condition indices 164

6.3.3 Body condition indices

A significant difference between Populations A and B was detected for females and males for some BCIs. The Ratio and Ratio.2 indices showed a significant difference among females, and Quetelet’s, Fulton’s K and OLS showed a significant difference for males. These results are summarised in Table 6-5. In this study 54% (n = 13) of individuals sampled in Population B had less than one gram (g) of renal fat removed from their kidney, 42% (n = 10) had less than five grams, and one male had 21.7g of renal fat. The mean (± se) KFI for females and males was 3.21 ± 0.88 (range = 0.21 and 9.70%), and 4.44 ± 2.45 (range = 0.19 and 23.04%), respectively.

Table 6-5 Summary statistics for each body condition index between two populations of eastern grey kangaroos (Macropus giganteus) for females and males.

Females Males Body Condition Index t-statistic p-value t-statistic p-value Ratio 3.269 0.005 1.694 0.107 Quetelet’s 1.950 0.061 2.151 0.045 Fulton’s K 1.653 0.110 3.191 0.006 Ratio.2 2.896 0.011 1.499 0.151 OLS 1.531 0.137 3.251 0.004

Given the significant differences in leg length and body weight measurements between the two populations for females, the OLS data have also been presented to show the population specific relationship between ln-transformed body weight (kg) and ln- transformed leg length (cm) (Figure 6-3(a) females, and 6-3(b) males); and ln- transformed body weight (kg) and ln-transformed pes length (cm) (Figure 6-4(a) females, and 6-4(b) males). These graphs show, especially for leg length in females, that there was a size bias between the populations, with population B having five females with an ln-leg length < 1.64. There were no females from population A within this leg length range. This has likely increased the emphasis of population B on calculation the OLS regression equation.

Chapter 6 Validation of body condition indices 165

(a) 1.60

1.50 ) 1.40

1.30

1.20 Log weight (kg weight Log

1.10

1.00 1.54 1.56 1.58 1.60 1.62 1.64 1.66 1.68 1.70 1.72 Log leg length (cm)

(b)

1.90 1.80 1.70 ) 1.60 1.50 1.40 1.30 Log weight (kg weight Log 1.20 1.10 1.00 1.60 1.65 1.70 1.75 1.80 1.85 Log leg length (cm)

Figure 6-3 Individual differences in the relationship between ln-transformed body weight (kg) and leg length (cm) for (a) females, and (b) males in Population A (‘good’ condition; ) and Population B (‘poor’ condition; ).

Chapter 6 Validation of body condition indices 166

(a)

1.60

1.50 ) 1.40

1.30

1.20 Log weight (kg weight Log

1.10

1.00 1.40 1.42 1.44 1.46 1.48 1.50 1.52 1.54 Log pes length (cm)

(b)

1.90 1.80 1.70 ) 1.60 1.50 1.40 1.30 Log weight (kg weight Log 1.20 1.10 1.00 1.40 1.45 1.50 1.55 1.60 Log pes length (cm)

Figure 6-4 Individual differences in the relationship between ln-transformed body weight (kg) and pes length (cm) for (a) females, and (b) males in Population A (‘good’ condition; ) and Population B (‘poor’ condition; ).

Chapter 6 Validation of body condition indices 167

6.3.4 Validation of body condition indices using blood parameters

This study aimed to validate five BCIs at the individual level in addition to the population level by assessing animals independently of the population, and examining whether there is a positive relationship between each BCI and the selected blood parameters. Validating these indices at an individual level would allow the identification of those individuals within a population that were in relatively good or poor condition. None of the five BCIs tested could detect a difference in condition between Population A and Population B for both males and females. Females showed no significant relationship between the serum biochemistry parameters and any BCI (p > 0.05). There was also no significant relationship detected between the haematology parameters and any BCIs (p > 0.05). In males, of the five biochemistry parameters tested, there was a significant relationship between Fulton’s K and CK (Wald chi- squared = 5.284, p = 0.022), but all other relationships were non-significant (p > 0.05). The GLM for haematology showed significant relationships between the ratio BCIs and haemoglobin (Ratio: Wald chi-squared = 18.971, p = 0.000; Quetelet’s: Wald chi- squared = 18.743, p = 0.000; Fulton’s K: Wald chi-squared = 11.982, p = 0.001; Ratio.2: Wald chi-squared = 22.122, p = 0.000). All other relationships were non- significant (p > 0.05). In females, of the serum biochemistry and haematology parameters, there was significant relationship between KFI and total protein (Wald chi- squared = 5.793, p = 0.016) and CK (Wald chi-squared = 4.009, p = 0.045). For males, there was a significant relationship between KFI and creatinine (Wald chi-squared = 6.713, p = 0.010) and total protein (Wald chi-squared = 5.864, p = 0.015). All other variables for both females and males were not significant (p > 0.05).

6.4 Discussion

In this study we sought to validate five commonly used BCIs using serum biochemistry and haematology parameters. There are four main conclusions that can be made from these results: (i) serum biochemistry and haematology were more reliable indicators of body condition at both an individual and population level; (ii) no one BCI calculated from the morphological data reliably predicted the condition of a population for both sexes; (iii) there was generally no positive association between any BCI and blood parameter; and (iv) currently used BCIs are not sensitive indicators of

Chapter 6 Validation of body condition indices 168 body condition at a population level for eastern grey kangaroos. Although both ratio and residual BCIs have been widely used (Bakker & Main 1980; Bercovitch et al. 2003; Green 2001; Viggers et al. 1998; Whitfield et al. 1999), they have rarely been applied to macropodids, and few studies have attempted to validate condition indices using serum biochemistry and haematology.

6.4.1 Serum biochemistry and haematology

The serum biochemistry and haematology showed individuals from Population A to be in a significantly better condition than Population B. Haematological reference ranges have been published for many macropodids (Clark 2004), however the variation among species highlights the need for species-specific data. The values for the serum biochemistry and haematology for the eastern grey kangaroo reported in Population A, but generally not Population B, were within the range reported for captive eastern grey kangaroos (ISIS 2002), although the values for ALP in this study were much lower (this study females: 308 ± 32 IU/L; males = 556 ± 67 IU/L; published: 1037 ± 775 IU/L) (ISIS 2002). It has been reported that ALP concentrations over 1000 IU/L are common in macropodids, but both populations in this study exhibited lower levels (Vogelnest & Portas 2008). Alkaline phosphastase (ALP) is produced in the liver, bones and intestines and elevated levels are linked to bone disease, growth and pregnancy (Vogelnest & Portas 2008). The significance of lower levels is unknown.

Consistent with our expectations, the pathology results showed Population B was in significantly poorer condition than Population A. The serum biochemistry and haematology of population B differed significantly from Population A for most parameters (Table 6-3(a) and 6-3(b)) and the results previously reported in captive eastern grey kangaroos (ISIS 2002). Urea evaluates kidney function and is a measure of protein breakdown in the body and increases may result from an increase in dietary protein, catabolic breakdown of tissue, and haemorrhage into the gastrointestinal tract (Meyer & Harvey 1998). Creatinine also evaluates kidney function and is a breakdown product of creatine, which is an important component of muscle. As the fluid flow rate though the kidney (glomerular filtration) decreases, creatinine levels rise. Albumin and globulin are the two main proteins circulating in the blood, and together these proteins are measured as total protein. Protein measurements can reflect nutritional state,

Chapter 6 Validation of body condition indices 169 kidney disease, liver disease, and many other conditions. The low levels of albumin, globulin and total protein levels in Population B could be attributed to conditions such as protein losing gastroeneteropathies (including diarrhoea), dehydration, kidney and liver dysfunction, disease, compromised immune function, malabsorption (inadequate absorption of nutrients from the intestinal tract) or malnutrition (Meyer & Harvey 1998). ALT and AST are enzymes found in the liver and the high levels of ALT and AST present in individuals in Population B are consistent with these ailments associated with low proteins. ALP, AST, ALT and CK are indicative of liver condition. AST is found in high concentration in heart muscle, liver cells, and skeletal muscle cells. CK measures creatine phosphokinase (CPK), and high AST and CK are indicative of muscle damage - heart, brain or muscle tissue.

The haematology results are consistent with the poor condition revealed by the serum biochemistry (Meyer & Harvey 1998). The low erythrocyte count, haemoglobin and haematocrit suggest the animals are suffering from anaemia, are in poor nutritional state, and have kidney and liver dysfunction (Clark 2004). White cell counts can fluctuate with a number of diseases and the low levels in Population B could be indicative of infection and/or disease. Overall, the differences in the blood parameters between Population A and B reflect the nutritional quality of food available and the lack of water availability for the wild Population B.

Sex differences were noted for most blood parameters in Population A, and only two in Population B. In Population A, kangaroo pellets form a significant component of their diet. These pellets have a relatively high protein content (14%) which may explain the lack of difference between females and males in total protein, albumin and globulin, all of which are linked to protein. The lack of sex differences in Population B may be a result of the overall poor condition of the population. However, no sex differences have been noted in haematological variables in other marsupials, for example agile wallabies (M. agilis) (Stirrat 2003) and southern brown bandicoots (Isoodon obesculus) (Wicks & Clark 2005).

Chapter 6 Validation of body condition indices 170

6.4.2 Body condition indices

Although BCIs could detect a significant difference between the two populations, no one condition index emerged as a consistent predictor. Two condition indices, Ratio and Ratio.2, indicated females in Population A as being in significantly better condition than females in Population B. However, for males, three different condition indices (Quetelet’s, Fulton’s K and OLS) detected a difference in body condition between population A and B. This was surprising as the populations selected for this study represent two extremes in body condition as reflected by the blood parameters. This suggests that the five BCIs used in this study may not be consistently reliable predictors of condition at a population level. Consistent with other studies, the KFI was a good indicator of poor condition in Population B (Spinage 1984; Takatsuki 2000; van Rooyen 1993; Vicente et al. 2007), but there was no comparable data for the good condition population A. Both females and males from Population B had very little kidney fat present, similar to those reported in other species, for example the mean KFI in a starving population of western grey kangaroos (M. fuliginosus) and red kangaroos (M. rufus) was 4.6% and 6.5% respectively (Shepherd 1987).

6.4.3 Relationship between blood parameters and body condition indices

Attempts to validate BCIs with serum biochemistry and haematology data were unsuccessful. For many years the accuracy, reliability, repeatability and representation of BCIs has been questioned (Bakker & Main 1980; Cone 1989; Green 2001; Schulte- Hostedde et al. 2001a; Schulte-Hostedde et al. 2005a). However, these BCIs are ecologically and evolutionarily important as they correlate with many aspects of vertebrate life history traits, and many of these traits are condition-dependent, for example testis size in small mammals (Schulte-Hostedde & Millar 2004; Schulte- Hostedde et al. 2005a), antlers in male red deer (Cervus elaphus) (Clutton-Brock et al. 1982) and horns of bighorn sheep (Ovis canadensis) (Festa-Bianchet et al. 2004). This study once again raises the question of the reliability of current body condition methodologies and their interpretation. This is an ongoing study that aims to collect more samples with the ultimate goal of devising a new non-destructive condition index that can be used in the field that is both accurate and repeatable.

Chapter 6 Validation of body condition indices 171

Blood parameters not only reflect body condition, but also the health status, and changes in nutritional state for an individual. Some studies have examined the relationship between haematocrit and body condition in birds (Cuervo et al. 2007; Johnson & Albrecht 1993; Sanchez-Guzman et al. 2004; Simon et al. 2004), but currently there is no consensus on the suitability of haematocrit as a predictor of body condition (Dawson & Bortolotti 1997a; Dawson & Bortolotti 1997b; Villegas et al. 2002) and this study found no such relationship. The underlying physiological reason for the lack of a significant relationship between the blood parameters and the BCIs is unclear at this stage.

The lack of relationship between the BCIs and blood parameters may be a result of small sample size, although the number of individuals sampled here was similar to that reported in other body condition studies (Schulte-Hostedde et al. 2001a; Viggers et al. 1998). Krebs & Singleton (1993) conducted a study comparing the body condition of two populations of house mice (Mus domesticus) and noted that a sample size of 15 was required for statistical power. This study aimed to sample at least 15 individuals of each sex in each population but was restricted by the difficulties associated with sampling populations that were not within a close proximity to laboratories to process the samples. Perhaps further confounding any sample size effect was the fact that there was a size bias between females sampled from populations A and B. The inclusion of smaller females from population B has likely increased the influence of population B on calculation the OLS regression equation, with follow-on effects on OLS residual calculation.

It is possible that variation in capture and sample collection methods between the two populations may have influenced the blood biochemistry and/or haematology values. Animals from population A were captured by chemical immobilisation, whilst animals in population B were shot. The stress associated with capture and the physiological effects of the immobilising agent may have influenced blood parameters in the ‘good’ condition animals relative to the ‘poor’ condition shot animals that would not have had the time to mount a physiological response to ‘capture’. Creatine kinase (CK) is one biochemical analyte that may be expected to change in response to capture stress. CK is a muscle enzyme and increases are associated with trauma and muscle damage.

Chapter 6 Validation of body condition indices 172

Levels of CK have been shown to rise to 2000 – 3000 IU/L following physical or chemical restraint as a rapid response to cardiac and muscle damage (Vogelnest & Portas 2008). However, the levels of CK for Population A were significantly lower (359 – 502 IU/L) than these reported indicating that animal capture had little influence on the circulating levels of CK in the blood of animals used in this study. Future sample collection will aim to test the effects of animal capture on blood parameters and to ensure a more even spread of size categories between ‘good’ and ‘poor’ condition populations.

6.4.4 Conclusions

The aim of this investigation was to validate five commonly used BCIs with serum biochemistry and haematology to provide a scientifically based rationale for decision making in wildlife management. Overall the serum biochemistry and haematology indicated that Population A was in significantly better condition than Population B. The biochemistry results indicate that Population B is suffering from malnutrition, dehydration, and chronic disease which are negatively impacting upon their liver and kidney function. No single condition index could reliably detect differences between Population A and B, nor was there a positive association between the blood parameters and BCIs. This study found that serum biochemistry and haematology are more reliable indicators of body condition in eastern grey kangaroos than BCIs calculated from morphological data which provided inconsistent results. The populations used in this study were in grossly different conditions and these BCIs could not reliably detect this. This raises the question of what should be used to determine body condition in animals, particularly when management decisions are influenced by such measures. However, using blood parameters is not the answer as sample collection is time- consuming, and expensive and limited to areas within close proximity to a laboratory for processing within 24 hours of collection. This study forms part of ongoing research continuing to validate BCIs with blood parameters, with the intention of developing a key criteria and sensitive condition index that is non-invasive and can be readily applied to wild populations on site.

Chapter 7 General discussion 173

Chapter 7 General Discussion: Marsupial genetics, reproduction and the implications for wildlife management

To ensure the continued survival of many species, active in-situ and ex-situ conservation and management must be undertaken. Wildlife management aims to remove or reduce threatening processes, and increase population number and size. Such measures are necessary to reduce the effects of deterministic and stochastic threats and so reduce the risk of extinction. Although many recommendations have been put forward to enhance species survival, most conservation efforts would benefit from a more thorough understanding of the biology and genetics of the target species. The aim of this thesis was to demonstrate the importance and benefits of having a clearer understanding of species biology and the usefulness of genetics in conservation biology and wildlife management, regardless of whether a species is threatened or overabundant.

The first step in the management of both wild and captive populations is to resolve taxonomic uncertainties. Incorrect identification can result in the denial of protection for some species and is important for species undergoing translocations and reintroductions to avoid unwanted introgression (Allendorf & Luikart 2007). Resolving taxonomic uncertainties is also essential for defining units of conservation within a species in order to protect distinct lineages and maintain the evolutionary potential of a species. There has been much debate on the best method for identifying ‘units’ of conservation below the species level (Crandall et al. 2000; Moritz 1994; Moritz 1999; Ryder 1986; Waples 1991), with most scientists now recognising that it Chapter 7 General discussion 174 is important to conserve the evolutionary potential of a population, through the conservation of Evolutionary Significant Units (ESU). While the criteria for designating ESUs are hotly contested, there is general acceptance that an ESU is a population or group of populations demonstrating significant divergence in allele frequency with restricted gene flow to other such units, as such it has unique evolutionary potential which is a product of previous evolutionary events (Fraser & Bernatchez 2001). Below the level of an ESU, is a Management Unit (MU). Management Units are considered to be the ecological components of ESUs (Moritz 1999), with the focus here on contemporary population structuring and short term monitoring, rather than historical factors (Fraser & Bernatchez 2001). The current study of the phylogenetic relationships within tammar wallabies (Chapter 4) has demonstrated that populations from SA and WA represent distinct evolutionary lineages, reflecting their historical isolation and thereby warrant being managed as separate ESUs pending additional research and possible taxonomic revision. Within WA, the Abrolhos and mainland populations showed significant genetic structuring in their allelic and haplotypic frequencies, sufficient to be preserved as separate MUs to conserve diversity and the evolutionary processes within the WA ESU (Chapter 4).

These findings illustrate how adaptive differences can arise under restricted gene flow and reproductive isolation (Crandall et al. 2000). This raises the question of whether these populations should be maintained in isolation as independently evolving units [to maintain their adaptive variability within each unit], or mixed with other populations to maintain genetic variability. Since all the Abrolhos tammar populations are inbred, subdivided and have no recent gene flow, they may benefit from the translocation of new individuals into the population to increase genetic diversity, reducing the impact of inbreeding and preventing further genetic deterioration (e.g., kinked tails). There are three potential options: (1) augment the Abrolhos populations with new individuals from mainland WA as they are all a single ESU. However, the Abrolhos tammar wallabies are morphologically and reproductively distinct from the mainland (both WA and SA) populations (Herbert et al. unpublished data). Since the mainland tammars are not adapted to the harsh island environmental conditions, they may not survive if introduced. As with the movement of any animals, there would be a potential risk of introducing diseases and/or pathogens into the island populations, which would Chapter 7 General discussion 175

be particularly susceptible considering their long-term isolation and lack of exposure. (2) Animals could be exchanged between EWI and WWI. Both populations are genetically distinct from one another (separate MUs) and an exchange would lead to an increase in genetic diversity in both populations and a reduction in inbreeding. Although it is not recommended that inbred or genetically depauperate populations be used to augment other populations in an effort to maximise genetic diversity as it can compromise their long-term evolutionary plasticity (Eldridge et al. 1999; Moritz 1999), both populations could benefit from the exchange of new individuals. EWI and WWI are most closely related to each other and therefore would be suitable candidates for the exchange of individuals. (3) NI could be augmented with individuals from WWI and EWI, transforming NI into a reservoir for a genetically diverse generic Abrolhos tammar population while maintaining the genetic integrity of both EWI and WWI (Chapter 4). Ultimately the option pursued depends on how these island populations are viewed from a management perspective and whether the priority is to retain their natural genetic integrity or use supplementation to increase their genetic diversity and fitness and so reduce their risk of extinction.

One of the long-term goals of conservation and management is to preserve sufficient genetic variation for species to retain their evolutionary potential (Allendorf & Luikart 2007). Genetic diversity is maintained through a balance of mutation, genetic drift, and natural selection (Frankham 1996). Genetic variation arises through mutation and is lost through drift in populations with a finite size, and natural selection may reduce or enhance the retention of diversity through balancing or diversifying selection (Frankham 1996). The rate of loss is will depend of the effective population size (Ne) and the number of generations a population is isolated (Frankham 1997). In the case of the Abrolhos Island tammar wallaby populations (Chapter 4), they have lost almost half the diversity present in the mainland WA population as a result of random genetic drift and the lack of recent gene flow. However, some variability is likely to have been replenished through mutation. The genetic data showed no evidence of migrants contributing to their variability. At some point, populations are expected to come into a balance of mutation and genetic drift (mutation-drift equilibrium) and the population will reach an equilibrium in genetic variation (Slatkin 1995). However, temporal fluctuations in the effective population size will disrupt this balance. Genetic theory Chapter 7 General discussion 176

also predicts that genetic variation within a population is related to populations and island size (Frankham 1997; Soulé 1976), and there was evidence for this trend in Abrolhos populations. The largest island (WWI) with the largest population had the highest genetic diversity, while the smallest island (NI) had the lowest level of diversity (Chapter 4).

Knowledge of the genetic variability within and between populations is important to assess the impact of founder effects, gene flow, genetic drift and bottlenecks (Chapters 2 and 4). Populations founded with a small number of individuals will suffer a loss of genetic diversity and inbreeding (Chapter 4) if there is restricted or no gene flow, as shown in the introduced NI tammar wallaby population (Chapter 4). Fragmented mainland and island populations draw many parallels and are of particular concern as they are more prone to the risk of extinction. The isolation of island populations exacerbates the effects of limited founder numbers though small population size, restricted gene flow, random genetic drift, genetic bottlenecks and inbreeding (Eldridge et al. 1999; Frankham 1997, 1998). Since islands are increasingly being used as refuges for threatened species, and a source of animals for conservation initiatives (Boessenkool et al. 2007; Courtenay & Friend 2003; Daltry et al. 2001; Eldridge et al. 2004; Rankmore et al. 2008), it is essential for conservation planning to take into consideration the number of founders and genetic variation for long-term planning.

Establishing populations of threatened species on islands may protect them from deterministic threats on the mainland, but will increase the impact of stochastic events such as random genetic drift, particularly if they are founded with a small number of individuals that may already be inbred due to population decline on the mainland. The tammar wallaby populations in the Abrolhos Islands provided a clear example of this. These populations have significantly lower genetic diversity than the mainland WA tammar wallaby population. Due to their isolation, there has been no recent gene flow between the populations, which has led to significant genetic differentiation and inbreeding (Chapter 4). In the case of the Abrolhos Island tammars, the potential to augment the genetic diversity on the islands (through the translocation of individuals) should be considered, given the current level of inbreeding and the likelihood of further erosion of genetic diversity in the future. In situations where insurance Chapter 7 General discussion 177

populations are marooned on islands (e.g. Gilbert’s potoroo), the Abrolhos Island data demonstrates the potential consequences of genetic isolation of populations, and leads to the suggestion that it would be beneficial to regularly exchange individuals between the introduced island and existing mainland populations to simulate natural gene flow and so assist in the maintenance of genetic diversity.

A growing number of captive breeding programs are being established to protect threatened species from extinctions and restore declining natural populations, (Frankham 2008; Seddon et al. 2007). It is recommended that captive breeding programs should be founded with at least 20 to 30 individuals to capture sufficient representation of genetic diversity from wild populations (Frankham et al. 2002), but many programs are founded with fewer animals than this. Small, closed populations that are founded with a small number of individuals will inevitably experience a loss of genetic diversity, inbreeding, the accumulation of deleterious alleles and genetic differentiation. However, regular gene flow (e.g., translocation and mating) into a population can assist in the maintenance of genetic diversity, and buffer against the negative effects from small founders, the loss of genetic diversity, inbreeding and population genetic differentiation, for both wild and captive populations. Two captive populations of the greater bilby were founded in 1997/1998, each with only seven individuals, but by simulating gene flow through the translocation of new individuals into each population, the level of genetic diversity has been maintained over several years, and contributed to the low levels of inbreeding observed in each captive colony (Chapter 2).

Captive breeding programs are often managed through pedigrees that are maintained in studbooks (Hedrick & Millar 1992). In some cases, pedigree information may be unknown or missing, for example the relationship between wild-caught founding individuals (Chapter 2). The genetic effects of establishing a population with a small number of individuals can be compounded if the founders are related or inbred (Chapter 4). Studbooks assist in minimising kinship among individuals, maintaining the representation of wild founder genetic diversity as well as providing a tool for assessing the genetic health of a population based on demographic data (Chapter 2). However, errors in studbook estimates of the ‘genetic health’ of a population can arise Chapter 7 General discussion 178

due to missing or erroneous pedigree data (Chapter 2). The calculation of kinship values for captive breeding programs depends on a population’s pedigree. These calculations are only effective when the relationships among individuals in a population are accurately known (Hedrick & Kalinowski 2000; Nielsen et al. 2007; Rudnick & Lacy 2008). Unfortunately all pedigrees have some level of uncertainty, especially since many were established during the early days of captive breeding when records were not always consistently maintained (Rudnick & Lacy 2008). In addition, most pedigrees have incomplete records associated with the founder individuals (i.e. they are usually assumed to be unrelated) used to establish the population (Rudnick & Lacy 2008). To improve the concordance between studbook and genetic estimates of diversity, institutions need to maintain more complete pedigree records. Validation of the ‘founder assumption’ of individuals being unrelated and not inbred (Ballou 1984), should be prioritised when there are limited resources available. Genetic analyses of parentage can help resolve questions regarding individual relationships within a population (Chapters 3 and 5), and verify pedigrees. Accurate parentage data is essential for wildlife managers when making decisions regarding the pairing of individuals for mating and the transfer of individuals among institutions. The widespread application of genetics has revealed inconsistencies in the presumed relationship between social organisation, parentage and mating system (Ambs et al. 1999; Coltman et al. 1999a; Issac 2005; Worthington Wilmer et al. 1999), transforming how mating systems are understood (Chapter 3).

Ideally, individuals selected for reintroduction should be physically healthy with a known high reproductive potential and high genetic variation (Frankham et al. 2002). In the absence of this information, individuals may be selected for release based on other traits such as body size in the greater bilby (Chapter 3). Much research in mammalian taxa has shown body size to be an important asset for males through increased reproductive success (Andersson 1994; Birkhead 2000), including the eastern grey kangaroo (Chapter 5), but this is not always the case (Chapter 3). Other traits can also influence male reproductive success, such as dominance status, testosterone concentrations and heterozygosity (Chapter 5). But in cryptic species such as the greater bilby, it can be difficult to determine what traits are associated with a high reproductive output (Chapter 3). In such cases, other parameters should be used to Chapter 7 General discussion 179

select males for reintroduction, such as characteristics that are likely to reflect the chances of successful survival post-release, for example health and body condition, age.

Maintaining genetic diversity usually requires maximising the effective population size (Frankham 1995b). Effective population size determines the level of genetic diversity maintained over time and is significantly influenced by a species mating system (Chapter 3). In a promiscuous mating system, for example the greater bilby (Chapter 3), more males participate in breeding, lowering the variance in male reproductive success which in turn, slows the rates of inbreeding, lowers the level of relatedness within the population and increases the effective population size (Frankham et al. 2002). The promiscuous mating system of the bilby has undoubtedly contributed to the maintenance of genetic diversity in free-ranging captive breeding programs for this species (Chapter 2). If the mating system is polygynous and there is a large reproductive skew with only a few males contributing to the gene pool, such as occurs in the eastern grey kangaroo (Chapter 5), there is greater potential for a decrease in genetic diversity in small isolated populations. In such cases, a potential strategy to maximise the effective population size in captive populations would be to manipulate mating patterns by using a specific number of sires (Oyama et al. 2007) to maximise founder representation. Although the eastern grey kangaroo is by no means endangered, these results highlight the importance of understanding the mating system of a species that is the subject of captive breeding.

Both within and between species, individuals differ widely in their reproductive success. This variation reflects differences in the allocation of resources such as time, energy and energy expenditure, to compete for life history functions such as growth and reproduction (Clutton-Brock 1988). This was demonstrated in two species with contrasting life history traits (Chapter 1, Table 1-1), the eastern grey kangaroo and greater bilby. For example, body size was an important determinant of male reproductive success in the eastern grey kangaroo (Chapter 5) but not bilbies (Chapter 3). The eastern grey kangaroo is a gregarious, territorial and social species (Ganslosser 1989; Jarman 1983) with a polygynous mating system (Chapter 5). They have a seasonal breeding season so female oestrus is more synchronous than in bilbies that Chapter 7 General discussion 180

can breed all year round. In a polygynous mating system multiple males compete for access to females, it can be predicted that selection will favour phenotypic adaptations that enhance a male’s ability to monopolise access to females or resources that attract females (Schulte-Hostedde et al. 2001b). In contrast, the bilby is a small, solitary species that has an overlap promiscuous mating system (Chapter 3) and can breed all year round. The spatial distribution is wide and varies with resource availability (Moseby & O'Donnell 2003), making it difficult for males to monopolise more than one female simultaneously. Consequently there may be selective pressure for high quality males that are skillful, efficient searchers as they are more likely to find a female when she becomes sexually receptive (Ims 1988), which may give the male a ‘prior possession advantage’ when another male arrives (Maynard Smith & Parker 1976; Maynard Smith & Price 1973).

The decision to cull a native marsupial population is often a controversial one. Such decisions are often based on the condition of the habitat, for example severe overgrazing or over browsing, or on the condition of individuals within the population, for example animals starving from a lack of sufficient forage. On North Island (WA), as with many kangaroo populations, a decision was made to cull the tammar wallaby population because of habitat destruction from overgrazing, and ultimately the animals were showing signs of poor condition and reduced reproductive output due to a lack of food availability. But how do wildlife managers assess body condition of wild populations? Chapter 6 shows that body condition indices are not consistent predictors of an individual’s health. The use of serum biochemistry and haematology provides a much more reliable indication of condition however this is expensive, time-consuming and not practical in remote locations. These results suggest there is currently no single measure that is adequate for reliably measuring body condition. Blood parameters and lethal sampling methods are the most reliable methods at the moment, however further sampling of different populations of different age structuring and condition may assist in developing a new method for calculating BCI that is reliable and non-invasive. As body condition plays an important role in management and conservation, further research and evaluation of the use of condition indices is required. Chapter 7 General discussion 181

In conclusion, genetics is a powerful tool for conservation and management, particularly with regard to behavioural, demographic and geographic information. Due to the rapid advances in genetic techniques, such methods are becoming more feasible to answer questions at a species, population and individual level. Genetic data can be used to assess phylogeography, introgression, to define management units; identifying source populations, as well as gaining insights into the historical demographic patterns of a population. Genetic data can also be used to elucidate mating systems, patterns of gene flow, inbreeding and effective population size. All of these factors provide important information about a species’ biology, and contribute information to their conservation and management. The research conducted in this thesis applied genetics to examine factors influencing genetic variation both within and between populations, identified source populations and the consequences of small founder effects, as well as patterns of parentage and mating systems. A broad range of biological data was also incorporated to compliment the genetic analyses including behavioural, morphological and physiological information to address a wide array of questions in the field of conservation genetics and wildlife management. It is hoped that this research will assist in efforts to more effectively manage wildlife populations and contribute to improved outcomes for overabundant species and the conservation of global diversity.

References 182

References

Abbott, I., and A. A. Burbridge. 1995. The occurrence of mammal species on the islands of Australia: a summary of existing knowledge. CALM Science 1:259- 324. Acevedo-Whitehouse, K., F. Gulland, D. Greig, and W. Amos. 2003. Disease susceptibility in California sea lions. Nature 422:35-35. Acevedo-Whitehouse, K., T. R. Spraker, E. Lyons, S. R. Melin, F. Gulland, R. L. Delong, and W. Amos. 2006. Contrasting effects of heterozygosity on survival and hookworm resistance in California sea lion pups. Molecular Ecology 15:1973-1982. Adderton Herbert, C. 2004. Long-acting contraceptives: a new tool to manage overabundant kangaroo populations in nature reserves and urban areas. Australian Mammalogy 26:67-74. Alexander, W. B. 1922. The vertebrate fauna of Houtman's Abrolhos (Abrolhos Islands), Western Australia. Journal of the Linnean Society of London. Zoology 34:457-486. Allendorf, F. W. 1986. Genetic drift and the loss of alleles versus heterozygosity. Zoo Biology 5:181-190. Allendorf, F. W., P. R. England, G. Luikart, P. A. Ritchie, and N. Ryman. 2008. Genetic effects of harvest on wild animal populations. Trends in Ecology and Evolution 23:327-337. Allendorf, F. W., and G. Luikart 2007. Conservation and the Genetics of Populations. Blackwell Publishing, Victoria, Australia. Altmann, S. A. 1962. A field study of the sociobiology of rhesus monkeys, Macaca mulatta. Annals of the New York Academy of Sciences 102:338-435. Ambs, S. M., D. J. Boness, W. D. Bowen, E. A. Perry, and R. C. Fleischer. 1999. Proximate factors associated with high levels of extraconsort fertilisation in polygynous grey seals. Animal Behaviour 58:527-535. Amos, W., J. Worthington Wilmer, K. Fullard, T. M. Burg, J. P. Croxall, D. Bloch, and T. Coulson. 2001a. The influence of parental relatedness on reproductive success. Proceedings of the Royal Society of London B: Biological Sciences 268:2021-2027.

References 183

Amos, W., J. Worthington Wilmer, and H. Kokko. 2001b. Do female grey seals select genetically diverse mates? Animal Behaviour 62:157-164. Andersson, M. 1994. Sexual Selection. Princeton University Press, New Jersey, USA. Araki, H., W. R. Ardren, E. Olsen, B. Cooper, and M. S. Blouin. 2007. Reproductive success of captive-bred steelhead trout in the wild: evaluation of three hatchery programs in the Hood river. Conservation Biology 21:181-190. Arnold, S. J., and M. J. Wade. 1984. On the measurement of natural and sexual selection: applications. Evolution 38:720-734. Atkinson, S. N., and M. A. Ramsay. 1995. The effects of prolonged fasting on the body composition and reproductive success of female polar bears (Ursus maritimus). Functional Ecology 9:559-567. Bakker, H. R., and A. R. Main. 1980. Condition, body composition and total body water estimation in the quokka, Setonix brachyurus (Macropodidae). Australian Journal of Zoology 28:395-406. Ballou, J. D. 1984. Strategies for maintaining genetic diversity through reproductive technology. Zoo Biology 3:311-323. Ballou, J. D., and R. C. Lacy. 1995. Identifying genetically important individuals for management of genetic variation in pedigreed populations. Pages 76-111 in J. D. Ballou, M. Gilpin, and T. Foose, editors. Population Management for Survival and Recovery. Columbia University Press, New York, USA. Barrett, G. M., K. Shimizu, M. Bardi, S. Asaba, and A. Mori. 2002. Endocrine correlates of rank, reproduction, and female-directed aggression in male Japanese macaques (Macaca fuscata). Hormones and Behaviour 42:85-96. Bartlett, J. M. S., and D. Stirling. 2003. A short history of the polymerase chain reaction. Methods in Molecular Biology 226:3-6. Bazin, E., S. Glemin, and N. Galtier. 2006. Population size does not influence mitochondrial genetic diversity in animals. Science 312:570-572. Beacham, J. L. 2003. Models of dominance hierarchy formation: effects of prior experience and intrinsic traits. Behaviour 140:1275-1303. Bean, K., W. Amos, P. P. Pomeroy, S. D. Twiss, T. N. Coulson, and I. L. Boyd. 2004. Patterns of parental relatedness and pup survival in the grey seal (Halichoerus grypus). Molecular Ecology 13:2365-2370.

References 184

Beldomenico, P. M., S. Telfer, S. Gebert, L. Lukomski, M. Bennett, and M. Begon. 2008. The dynamics of health in wild field vole populations: a haematological perspective. Journal of Animal Ecology 77:984-997. Berard, J. D., P. Nürnberg, J. T. Epplen, and J. Schmidtke. 1993. Male rank, reproductive behaviour, and reproductive success in free-ranging rhesus macaques. Primates 34:481-489. Bercovitch, F. B. 1991. Social stratification, social strategies, and reproductive success in primates. Ethology and Sociobiology 12:315-333. Bercovitch, F. B., and P. Nürnberg. 1996. Socioendocrine and morphological correlates of paternity in rhesus macaques (Macaca mulatta). Journal of Reproduction and Fertility 107:59-68. Bercovitch, F. B., A. Widdig, A. Trefilov, M. J. Kessler, J. D. Berard, J. Schmidtke, P. Nürnberg, and M. Krawczak. 2003. A longitudinal study of age-specific reproductive output and body condition among male rhesus macaques, Macaca mulatta. Naturwissenschaften 90:309-312. Bergman, T. J., J. C. Beehner, D. L. Cheney, R. M. Seyfarth, and P. L. Whitten. 2006. Interactions in male baboons: the importance of both males' testosterone. Behavioural Ecology and Sociobiology 59:480-489. Birkhead, T. R. 2000. Promiscuity: An Evolutionary History of Sperm Competition and Sexual Conflict. Faber and Faber Limited, London, UK. Birkhead, T. R., and A. P. Møller 1998. Sperm Competition and Sexual Selection. Academic Press, London, UK. Blows, M. W., and R. Brooks. 2003. Measuring nonlinear selection. The American Naturalist 162:815-820. Boessenkool, S., S. Taylor, C. Tepolt, J. Komdeur, and I. Jamieson. 2007. Large mainland populations of South Island robins retain greater genetic diversity than offshore island refuges. Conservation Genetics 8:705-714. BOM. 2008. National Tidal Centre. Bureau of Meterology http://www.bom.gov.au/oceanography/tides/. Bowyer, J. C., G. R. Newell, and M. D. B. Eldridge. 2002. Genetic effects of habitat contraction on Lumholtz's tree-kangaroo (Dendrolagus lumholtzi) in the Australian Wet Tropics. Conservation Genetics 3:61-69.

References 185

Brooke, M. d. L., S. H. M. Butchart, S. T. Garnett, G. M. Crowley, N. B. Mantilla- Beniers, and A. J. Stattersfield. 2008. Rates of movement of threatened bird species between IUCN Red List categories and toward extinction. Conservation Biology 22:417-427. Brooks, L. D. 2003. SNPs: why do we care? Single Nucleotide Polymorphisms: Methods in Molecular Biology:1-14. Bryant, E. H., V. L. Backus, M. E. Clark, and D. H. Reed. 1999. Experimental tests of captive breeding for endangered species. Conservation Biology 13:1487-1496. Burton, C. 2002. Microsatellite analysis of multiple paternity and male reproductive success in the promiscuous snowshoe hare. Canadian Journal of Zoology 80:1948-1956. Cabe, P. R. 1998. The effects of founding bottlenecks on the genetic variation in the European starling (Sturnus vulgaris) in North America. Heredity 80:519-525. Calaby, J. H., and G. Grigg. 1989. Changes in macropodid communities and populations in the past 200 years, and the future. Pages 813-820 in G. Grigg, P. J. Jarman, and I. D. Hume, editors. Kangaroos, Wallabies, and Rat-Kangaroos. Surrey Beatty and Sons Sydney, Australia. CALM. 1999. Draft Western Shield Fauna Recovery Program – Strategic Plan (July 1999-June 2004) in W. A. CALM, editor. Department of Conservation and Land Management. Carlson, A. A., A. J. Young, A. F. Russell, N. C. Bennett, A. S. McNeilly, and T. Clutton-Brock. 2004. Hormonal correlates of dominance in meerkats (Suricata suricatta). Hormones and Behaviour 46:141-150. Caughley, G. 1962. The comparative ecology of the red and grey kangaroo, MSc thesis. University of Sydney, Sydney, Australia. Caughley, G. 1987. Introduction to the sheep rangelands. Pages 1-13 in G. Caughley, N. Shepherd, and J. Short, editors. Kangaroos: Their Ecology and Management in the Sheep Rangelands of Australia. Cambridge University Press, Cambridge, UK. Caughley, G. 1994. Directions in conservation biology. Journal of Animal Ecology 63:215-244. Caughley, G., and A. Gunn 1996. Conservation Biology in Theory and Practice. Blackwell Science, Massachusetts, USA.

References 186

Ceballos, G., and P. R. Ehrlich. 2002. Mammal population losses and the extinction crisis Science 296:904-907. Charmantier, A., and B. C. Sheldon. 2006. Testing genetic models of mate choice evolution in the wild. Trends in Ecology and Evolution 21:417-419. Charpentier, M., P. Peignot, M. Hossaert-McKey, O. Gimenez, J. M. Setchell, and E. J. Wickings. 2005. Constraints on control: factors influencing reproductive success in male mandrills (Mandrillus sphinx). Behavioural Ecology 16:614- 623. Clark, F. M., and C. G. Faulkes. 1998. Hormonal and behavioural correlates of male dominance and reproductive status in captive colonies of the naked mole-rat, Heterocephalus glaber. Proceedings of the Royal Society of London B: Biological Sciences 265:1391-1399. Clark, P. 2004. Haematology of Australian Mammals. CSIRO publishing, Collingwood, Australia. Clinchy, M., A. C. Taylor, L. Y. Zanette, C. J. Krebs, and P. Jarman. 2004. Body size, age and paternity in common brushtail possums (Trichosurus vulpecula). Molecular Ecology 13:195-202. Clutton-Brock, T. H. 1988. Reproductive Success. Studies of Individual Variation in Contrasting Breeding Systems. University of Chicago Press, Chicago, USA. Clutton-Brock, T. H. 1989. Mammalian mating systems. Proceedings of the Royal Society of London B: Biological Sciences 236:339-372. Clutton-Brock, T. H., D. Green, M. Hiraiwa-Hasegawa, and S. D. Albon. 1988. Passing the buck: resource defence, lek breeding and mate choice in fallow deer. Behavioural Ecology and Sociobiology 23:281-296. Clutton-Brock, T. H., F. E. Guinness, and S. D. Albon 1982. Red Deer: Behaviour and Ecology of Two Sexes. University of Chicago Press, Chicago, USA. Clutton-Brock, T. H., and K. Isvaran. 2006. Paternity loss in contrasting mammalian societies. Biology Letters 2:513-516. Coltman, D. W., D. R. Bancroft, A. Robertson, J. A. Smith, and T. H. Clutton-Brock. 1999a. Male reproductive success in a promiscuous mammal: behavioural estimates compared with genetic paternity. Molecular Ecology 8:1199-1209.

References 187

Coltman, D. W., W. D. Bowen, and J. M. Wright. 1998. Male mating success in an aquatically mating pinniped, the harbour seal (Phoca vitulina), assessed by microsatellite DNA markers. Molecular Ecology 7:627-638. Coltman, D. W., W. D. Bowen, and J. M. Wright. 1999b. A multivariate analysis of phenotype and paternity in male harbor seals, Phoca vitulina, at Sable Island, Nova Scotia. Behavioural Ecology 10:169-177. Coltman, D. W., M. Festa-Bianchet, J. T. Jorgenson, and C. Strobeck. 2001. Age- dependent sexual selection in bighorn rams. Proceedings of the Royal Society of London B: Biological Sciences 269:165-172. Coltman, D. W., P. O'Donoghue, J. T. Jorgenson, J. T. Hogg, C. Strobeck, and M. Festa-Bianchet. 2003. Undesirable evolutionary consequences of trophy hunting. Nature 426:655-658. Coltman, D. W., J. G. Pilkington, J. A. Smith, and J. M. Pemberton. 1999c. Parasite- mediated selection against inbred soay sheep in a free-living, island population. Evolution 53:1259-1267. Cone, R. S. 1989. The need to reconsider the use of condition indices in fishery science. Transactions of the American Fisheries Society 118:510-514. Cooper, D. W., and C. A. Herbert. 2001. Genetics, biotechnology and population management of over-abundant mammalian wildlife in Australasia. Reproduction, Fertility and Development 13:451-458. Cornuet, J. M., and G. Luikart. 1996. Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144:2001-2014. Coulson, G. 1998. Management of overabundant macropods - are there conservation benefits? Pages 37-48 in A. Austin, editor. Managing Maruspial Abundance for Conservation Benefits. Cooperative Research Centre for the Conservation and Management of Marsupials, Sydney, Australia. Coulson, G. 2008. Eastern Grey Kangaroo. Pages 335-338 in S. Van Dyck, and R. Strahan, editors. The Mammals of Australia. Reed New Holland, Sydney, Australia. Coulson, T. N., J. M. Pemberton, S. D. Albon, M. Beaumont, T. C. Marshall, J. Slate, F. E. Guinness, and T. H. Clutton-Brock. 1998. Microsatellites reveal heterosis

References 188

in red deer. Proceedings of the Royal Society B: Biological Sciences 265:489- 495. Courtenay, J., and T. Friend. 2003. Gilbert's potoroo Recovery Plan July 2003 - June 2008 in W. A. Department of Environment and Conservation (DEC), editor. Department of Environment and Conservation. Crandall, K. A., O. R. P. Bininda-Emonds, G. M. Mace, and R. K. Wayne. 2000. Considering evolutionary processes in conservation biology. Trends in Ecology and Evolution 15:290-295. Croft, D. B. 1981. Behaviour of the red kangaroo, Macropus rufus (Desmarest, 1822) in north-western New South Wales. Australian Mammalogy 5:5-13. Cuervo, J. J., A. P. Møller, and F. de Lope. 2007. Haematocrit is weakly related to condition in nestling Barn Swallows Hirundo rustica. Ibis 149:128-134. Dallas, J. F. 1992. Estimation of microsatellite mutation rates in recombinant inbred strains of mouse. Mammalian Genome 3:452-456. Daltry, J. C., Q. Blooxam, G. Cooper, M. L. Day, K. Lindsay, and B. E. Smith. 2001. Five years of conserving the 'world's rarest snake', the Antiguan racer Alsophis antiguae. Oryx 35:119-127. Darwin, C. 1859. On the Origin of Species by Means of Natural Selection. John Murray, London, UK. Darwin, C. 1883. The Variation of Animals and Plants Under Domestication. D. Appleton and Co, New York, USA. Daugherty, C. H., A. Cree, J. M. Hay, and M. B. Thompson. 1990. Neglected taxonomy and continuing extinctions of tuatara (Spenodon). Nature 347:177- 179. Dawson, R. D., and G. R. Bortolotti. 1997a. Are avian hematocrits indicative of condition? American kestrels as a model. Journal of Wildlife Management 61:1297-1306. Dawson, R. D., and G. R. Bortolotti. 1997b. Variation in hematocrit and total plasma proteins of nestling American kestrels (Falco sparverius) in the wild. Comparative Biochemistry and Physiology - Part A: Molecular & Integrative Physiology 117:383-390. DEC. 2004. NatureBase: Bilby, Macrotis lagotis (Reid 1837). Department of Environment and Conservation, Western Australia, Perth, Western Australia.

References 189

DEH. 2004. Translocation proposal: re-introduction of mainland SA tammar wallaby to Innes National Park, Adelaide, South Australia in Department of Environment and Heritage. (DEH), editor. Government of South Australia, Adelaide, South Australia. DEH. 2007. The Kangaroo Conservation and Management Plan for South Australia 2008 - 2012 in Department of Environment and Heritage. (DEH), editor. Government of South Australia, Adelaide, South Australia. Deutsch, C. J., M. P. Haley, and B. J. Le Boeuf. 1990. Reproductive effort of male northern elephant seals: estimates from weight loss. Canadian Journal of Zoology 68:2580-2593. Dewsbury, D. A. 1982. Dominance rank, copulatory behaviour, and differential reproduction. The Quarterly Review of Biology 57:135-159. Dickinson, H. C., and J. E. Fa. 2000. Abundance, demographics and body condition of a translocated population of St Lucia whiptail lizards (Cnemidophorus vanzoi). Journal of Zoology 251:187-197. Dixson, A. F. 1998. Primate Sexuality: Comparative Studies of the Prosimians, Monkeys, Apes and Human Beings. Oxford University Press, Oxford, UK. Dlugosch, K. M., and I. M. Parker. 2008. Founding events in species invasions: genetic variation, adaptive evolution, and the role of multiple introductions. Molecular Ecology 17:431-449. Dobson, F. S., and W. T. Jones. 1985. Multiple causes of dispersal. American Naturalist 126:855-858. Dobson, F. S., and G. R. Michener. 1995. Maternal traits and reproduction in Richardson's ground squirrels. Ecology 76:851-862. Drake-Brockman, H. 1963. Voyage to Disaster. Angus and Robertson, Sydney, Australia. Eberhard, W. G. 1996. Female Control: Sexual Selection by Female Cryptic Choice. Princeton University Press, New Jersey, USA. Edwards, E. J., C. J. Still, and M. J. Donoghue. 2007. The relevance of phylogeny to studies of global change. Trends in Ecology and Evolution 22:243-249. Edwards, S. V., S. B. Kingan, J. D. Calkins, C. N. Balakrishnan, W. B. Jennings, W. J. Swanson, and M. D. Sorenson. 2005. Speciation in birds: genes, geography,

References 190

and sexual selection. Proceedings of the National Academy of Sciences 102:6550-6557. Eisenberg, J. F. 1966. The social organisation of mammals. Handbuch der Zoologie 10:1-92. Eknoyan, G. 2008. Adolphe Quetelet (1796 1874) the average man and indices of obesity. Nephrology Dialysis Transplantation 23:47-51. Eldridge, M. D. B., J. M. King, A. K. Loupis, P. B. S. Spencer, A. C. Taylor, L. C. Pope, and G. P. Hall. 1999. Unprecedented low levels of genetic variation and inbreeding depression in an island population of the black-footed rock-wallaby. Conservation Biology 13:531-541. Eldridge, M. D. B., J. E. Kinnear, K. R. Zenger, L. M. McKenzie, and P. B. S. Spencer. 2004. Genetic diversity in remnant mainland and "pristine" island populations of three endemic Australian macropodids (Marsupialia): Macropus eugenii, Lagorchestes hirsutus and Petrogale lateralis. Conservation Genetics 5:325-338. Ellegren, H. 1995. Mutation rates at porcine microsatellite loci. Mammalian Genome 6:376-377. Ellis, L. 1995. Dominance and reproductive success among nonhuman animals: a cross-species comparison. Ethology and Sociobiology 16:257-333. Emlen, S. T., and L. W. Oring. 1977. Ecology, sexual selection, and the evolution of mating systems. Science 197:215-223. Engelhardt, A., M. Heistermann, J. Hodges, P. Nürnberg, and C. Niemitz. 2006. Determinants of male reproductive success in wild long-tailed macaques (Macaca fascicularis) — male monopolisation, female mate choice or post- copulatory mechanisms? Behavioural Ecology and Sociobiology 59:740-752. England, P. R., and G. H. R. Osler. 2001. GENELOSS: A computer program for simulating the effects of population bottlenecks on genetic diversity. Molecular Ecology Notes 1:111-113. Escudero, A., J. M. Iriondo, and M. E. Torres. 2003. Spatial analysis of genetic diversity as a tool for plant conservation. Biological Conservation 113:351-365. Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14:2611-2620.

References 191

Ewen, K. R., P. D. Temple-Smith, D. K. Bowden, J. Marinopoulos, M. B. Renfree, and H. Yan. 1993. DNA fingerprinting in relation to male dominance and paternity in a captive colony of tammar wallabies (Macropus eugneii). Journal of Reproduction and Fertility 99:33-37. Excoffier, L., G. Laval, and S. Schneider. 2005. ARLEQUIN ver. 3.0: An integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1:47-50. Fairall, N. 1985. Manipulation of age and sex-ratios to optimise production from impala Aepyceros melampus populations. South African Journal of Wildlife Research 15:85-88. Fenberg, P. B., and K. Roy. 2007. Ecological and evolutionary consequences of size- selective harvesting: how much do we know? Molecular Ecology 17:209-220. Ferreira, Á. G. A., and W. Amos. 2006. Inbreeding depression and multiple regions showing heterozygote advantage in Drosophila melanogaster exposed to stress. Molecular Ecology 15:3885-3893. Festa-Bianchet, M., D. W. Coltman, L. Turelli, and J. T. Jorgenson. 2004. Relative allocation to horn and body growth in bighorn rams varies with resource availability. Behavioural Ecology 15:305-312. Ficetola, G. F., A. Bonin, and C. Miaud. 2008. Population genetics reveals origin and number of founders in a biological invasion. Molecular Ecology 17:773-782. Finnegan, L., C. Edwards, and J. Rochford. 2008. Origin of, and conservation units in, the Irish red squirrel (Sciurus vulgaris) population. Conservation Genetics 9:1099-1109. Fisher, D. O., and A. Cockburn. 2005. The large-male advantage in brown antechinuses: female choice, male dominance, and delayed male death. Behavioural Ecology:164-171. Fisher, D. O., and M. C. Lara. 1999. Effects of body size and home range on access to mates and paternity in male bridled nailtail wallabies. Animal Behaviour 58:121-130. Frankel, O. H. 1974. Genetic conservation: our evolutionary responsibility. Genetics 78:53-65. Frankel, O. H., and M. E. Soulé 1981. Conservation and Evolution. Cambridge University Press, Cambridge, UK.

References 192

Frankham, R. 1995a. Conservation genetics. Annual Review of Genetics 29:305-327. Frankham, R. 1995b. Effective population size/adult population size ratios in wildlife: a review. Genetical Research 66:95-107. Frankham, R. 1996. Relationship of genetic variation to population size in wildlife. Conservation Biology 10:1500-1508. Frankham, R. 1997. Do island populations have less genetic variation than mainland populations? Heredity 78:311-327. Frankham, R. 1998. Inbreeding and extinction: island populations. Conservation Biology 12:665-675. Frankham, R. 2008. Genetic adaptation to captivity in species conservation programs. Molecular Ecology 17:325-333. Frankham, R., J. D. Ballou, and D. A. Briscoe 2002. Introduction to Conservation Genetics. Cambridge University Press, Cambridge, UK. Fraser, D. J., and L. Bernatchez. 2001. Adaptive evolutionary conservation: towards a unified concept for defining conservation units. Molecular Ecology 10:2741- 2752. Fraser, D. J., M. M. Hansen, S. Østergaard, N. Tessier, M. Legault, and L. Bernatchez. 2007. Comparative estimation of effective population sizes and temporal gene flow in two contrasting population systems. Molecular Ecology 16:3866-3889. Freeman-Gallant, C. R., M. Meguerdichian, N. T. Wheelwright, and S. V. Sollecito. 2003. Social pairing and female mating fidelity predicted by restriction fragment length polymorphism similarity at the major histocompatibility complex in a songbird. Molecular Ecology 12:3077-3083. Fumagalli, L., L. C. Pope, P. Taberlet, and C. Moritz. 1997. Versatile primers for the amplification of the mitochondrial DNA control region in marsupials. Molecular Ecology 6:1199-1201. Ganslosser, U. 1989. Agonistic behaviour in Macropodoids - a review. Pages 475-503 in G. Grigg, P. Jarman, and I. Hume, editors. Kangaroos, Wallabies and Rat- Kangaroos. Surrey Beatty and Sons Pty Limited, Sydney, Australia. Garrott, R. A., P. J. White, and C. A. Vanderbilt White. 1993. Overabundance: an issue for conservationists? Conservation Biology 7:946-949. Gaston, K. J. 2000. Abundance-occupancy relationships. Journal of Applied Ecology 37:39-59.

References 193

Gaston, K. J., and R. A. Fuller. 2007. Commonness, population depletion and conservation biology. Trends in Ecology and Evolution 23 14-19. Gautschi, B., J. P. Muller, B. Schmid, and J. A. Shykoff. 2003. Effective number of breeders and maintenance of genetic diversity in the captive bearded vulture population. Heredity 91:9-16. Geoscience Australia. 2005. http://www.ga.gov.au/ in A. Government, editor. Australian Government. Gleeson, D., M. W. Blows, and I. P. F. Owens. 2005. Genetic covariance between indices of body condition and immunocompetence in a passerine bird. BMC Evolutionary Biology 5:61-70. Goodman, D. 2005. Selection equilibrium for hatchery and wild spawning fitness in integrated breeding programs. Canadian Journal of Fisheries and Aquatic Science 62:374-389. Gottelli, D., C. Sillero-Zubiri, G. D. Applebaum, M. S. Roy, D. J. Girman, J. Garcia- Moreno, E. A. Ostrander, and R. K. Wayne. 1994. Molecular genetics of the most endangered canid: the Ethiopian wolf Canis simensis. Molecular Ecology 3:301-312. Goudet, J. 2001. FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3). Available from http://www.unil.ch/izea/softwares/fstat.html. Updated from Goudet, J. (1995). FSTAT v-1.2. A computer program to calculate F-statistics. Journal of Heredity 86:485-486. Green, A. J. 2001. Mass/length residuals: measures of body condition or generators of spurious results? Ecology 82:1473-1483. Guinet, C., J. P. Roux, M. Bonnet, and V. Mison. 1998. Effect of body size, body mass, and body condition on reproduction of female South African fur seals (Arctocephalus pusillus) in Nambia. Canadian Journal of Zoology 76:1418- 1424. Haig, S. M., J. D. Ballou, and N. J. Casna. 1994. Identification of kin structure among Guam Rail founders: a comparison of pedigrees and DNA profiles. Molecular Ecology 3:109-119. Hamilton, W. D., and M. Zuk. 1982. Heritable true fitness and bright birds: a role for parasites? Science 218:384-387.

References 194

Hansson, B., H. Westerdahl, D. Hasselquist, M. Åkesson, and S. Bensch. 2004. Does linkage disequilibrium generate heterozygosity-fitness correlations in great reed warblers. Evolution 58:870-879. Harris, H. 1966. Enzyme polymorphism in man. Proceedings of the Royal Society B: Biological Sciences 164:298-310. Harris, R. B., W. A. Wall, and F. W. Allendorf. 2002. Genetic consequences of hunting: what do we know and what should we do? Wildlife Society Bulletin 30:634-643. Hasegawa, M., H. Kishino, and T. Yano. 1985. Dating of the human–ape splitting by molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22:160-174. Heath, D. D., J. W. Heath, C. A. Bryden, R. M. Johnson, and C. W. Fox. 2003. Rapid evolution of egg size in captive salmon. Science 299:1738-1740. Hedrick, P., R. Fredrickson, and H. Ellegren. 2001a. Evalution of d2, a microsatellite measure of inbreeding and outbreeding, in wolves with a known pedigree. Evolution 55:1256-1260. Hedrick, P. W. 1995. Gene flow and genetic restoration: the Florida panther as a case study. Conservation Biology 9:996-1007. Hedrick, P. W. 2000. Genetics of Populations. Jones and Bartlett, Boston, USA. Hedrick, P. W. 2001. Conservation genetics: where are we now? Trends in Ecology and Evolution 16:629-636. Hedrick, P. W., G. A. Gutierrez-Espeleta, and R. N. Lee. 2001b. Founder effect in an island population of bighorn sheep. Molecular Ecology 10:851-857. Hedrick, P. W., and J. S. Millar. 1992. Conservation genetics techniques and fundamentals. Ecological Applications 2:30-46. Hedrick, P. W., K. M. Parker, and R. N. Lee. 2001c. Using microsatellite and MHC variation to identify species, ESUs, and MUs in the endangered Sonoran topminnow. Molecular Ecology 10:1399-1412. Henderson, I. G., and P. J. B. Hart. 1995. Dominance, food acquisition and reproductive success in a monogamous passerine: the jackdaw Corvus monedula. Journal of Avian Biology 26:217-224. Herbert, C. A. 2007. From the urban fringe to the Abrolhos Islands: management challenges of burgeoning marsupial populations. Pages 129-141 in D. Lunney,

References 195

P. Eby, P. Hutchings, and S. Burgin, editors. Pest or Guest: The Zoology of Overabundance. Royal Zoological Society New South Wales, Mosman, Australia. Herbert, C. A., D. C. Eckery, T. E. Trigg, and D. W. Cooper. 2007. Chronic treatment of male tammar wallabies with deslorelin implants during pouch life: effects on development, puberty, and reproduction in adulthood. Biology of Reproduction 76:1054-1061. Hesperian. 2007. Satellite image of the Wallabi Group of the Houtman Abrolhos, including the outlying North Island. NASA World Wind using Landsat 7 data. Hinds, L. A., W. E. Poole, H. Tyndale-Biscoe, R. A. H. van Oorschot, and D. W. Cooper. 1990. Reproductive biology and the potential for genetic studies in the tammar wallaby, Macropus eugenii. Australian Journal of Zoology 37:223-234. Hobbs, R. H., and H. A. Mooney. 1998. Broadening the extinction debate: population depletions and additions in California and Western Australia. Conservation Biology 12:271-283. Hodges, K. E., C. I. Stefan, and E. A. Gillis. 1999. Does body condition affect fecundity in a cyclic population of snowshoe hares? Canadian Journal of Zoology 77:1-6. Hoelzel, A. R., B. J. Le Boeuf, J. Reiter, and C. Campagna. 1999. Alpha-male paternity in elephant seals. Behavioural Ecology and Sociobiology 46:298-306. Hoffman, J. I., I. L. Boyd, and W. Amos. 2004. Exploring the relationship between parental relatedness and male reproductive success in the Antarctic fur seal Arctocephalus gazella. Evolution 58:2087-2099. Holleley, C. E., C. R. Dickman, M. S. Crowther, and B. P. Oldroyd. 2006. Size breeds success: multiple paternity, multivariate selection and male semelparity in a small marsupial, Antechinus stuartii. Molecular Ecology 15:3439-3448. Hood, G. M. 2008. PopTools version 3.0.3. Available from: http://www.cse.csiro.au/poptools. Hoogland, J. L., and D. W. Foltz. 1982. Variance in male and female reproductive success in a harem-polygynous mammal, the black-tailed prairie dog (Sciuridae: Cynomys ludovicianus). Behavioural Ecology and Sociobiology 11:155-163.

References 196

Hossler, R. J., J. B. McAninch, and J. D. Harder. 1994. Maternal denning behaviour and survival of juveniles in opossums in southeastern New York. Journal of Mammalogy 75:60-70. Huelsenbeck, J. P., and F. Ronquist. 2001. MRBAYES: Bayesian inference of phylogenetic trees Bioinfomatics 17:754-755. Hughes, C. 1998. Integrating molecular techniques with field methods in studies of social behaviour: a revolution results. Ecology 79:383-399. Humphreys, W. F., A. J. Bradley, R. A. How, and J. L. Barnett. 1984. Indices of condition of phalanger populations: a review. Pages 59-77 in A. P. Smith, and I. D. Hume, editors. Possums and Gliders. Australian Mammal Society, Sydney, Australia. Hynes, E. F., C. D. Rudd, P. D. Temple-Smith, G. Sofronidis, D. Paris, G. Shaw, and M. B. Renfree. 2005. Mating sequence, dominance and paternity success in captive male tammar wallabies. Reproduction 130:123-130. Ingvarsson, P. K. 2003. Lone wolf to the rescue. Nature 420:472. Ingvarsson, P. K., and M. C. Whitlock. 2000. Heterosis increases the effective migration rate. Proceedings of the Royal Society B: Biological Sciences 267:1321-1326. ISIS. 1992. SPARKS (Single Population Analysis and Records Keeping System). International Species Information System. ISIS, Minnesota, USA. ISIS. 2002. Reference Ranges for Physiological Values in Captive Wildlife. International Species Information System. Eagan, Minnesota, USA. Issac, J. L. 2005. Potential causes and life-history consequences of sexual size dimorphism in mammals. Mammal Review 35:101-115. IUCN. 2006. 2006 IUCN Red List of Threatened Species www.iucnredlist.org IUCN. 2008. 2008 Red List summary statistics http://cmsdata.iucn.org/downloads/2008rl_stats_tables_all.xls. Jakob, E. M., S. D. Marshall, and G. W. Uetz. 1996. Estimating fitness: a comparison of body condition indices. Oikos 77:61-67. Jarman, P. J. 1983. Mating system and sexual dimorphism in large, terrestrial mammalian herbivores. Biological Reviews of the Cambridge Philosophical Society 58:485-520.

References 197

Jarman, P. J. 1989. Sexual dimorphism in Macropodoidea. Pages 433-447 in G. Grigg, P. J. Jarman, and I. D. Hume, editors. Kangaroos, Wallabies, and Rat- Kangaroos. Surrey Beatty and Sons, Sydney, Australia. Jarman, P. J. 1991. Social behaviour and organisation in the Macropodoidea. Advances in the Study of Behaviour 20:1-37. Jarman, P. J., and C. J. Southwell. 1986. Grouping, associations and reproductive strategies in eastern grey kangaroos. Pages 399-430 in P. I. Rubenstein, and R. W. Wrangham, editors. Ecological Aspects of the Social Evolution. Princeton University Press, New Jersey, USA. Jarne, P., and J. L. Logoda. 1996. Microsatellites, from molecules to populations and back. Trends in Ecology and Evolution 11:424-429. Jeffreys, A. J., V. Wilson, and S. L. Thein. 1985a. Hyper-variable "minisatellite" regions in human DNA. Nature 316:67-72. Jeffreys, A. J., V. Wilson, and S. L. Thein. 1985b. Individual-specific "fingerprints" of human DNA. Nature 316:76-78. Jiang, P., Q. Lang, S. Fang, P. Ding, and L. Chen. 2005. A genetic diversity comparison between captive individuals and wild individuals of Elliot’s Pheasant (Syrmaticus ellioti) using mitochondrial DNA. Journal of Zhejiang University Science B 6:413-417. Johnson, C. N. 1989. Social interactions and reproductive tactics in red-necked wallabies (Macropus rufogriseus banksianus). Journal of Zoology 217:267- 280. Johnson, C. N., and K. A. Johnson. 1983. Behaviour of the bilby, Macrotis lagotis (Reid), (Marsupialia, Thylacomyidae) in captivity. Australian Wildlife Research 10:77-87. Johnson, K. A. 2002. Bilbies. Pages 186-190 in R. Strahan, editor. The Mammals of Australia. Reed New Holland, Sydney, Australia. Johnson, L. S., and D. J. Albrecht. 1993. Effects of hematophagus ectoparasites on nestling House Wrens, Troglodytes aedon: who pays the cost of parasitism? Oikos 66:255-262. Johnson, M. S. 1988. Founder effects and geographic variation in the land snail Theba pisana. Heredity 61:133-142.

References 198

Johnstone, R. A. 2000. Models of reproductive skew: a review and synthesis. Ethology 106:5-26. Jones, K. L., T. C. Glenn, R. C. Lacy, J. R. Pierce, N. Unruh, C. M. Mirande, and F. Chavez-Ramirez. 2002. Refining the whooping crane studbook by incorporating microsatellite DNA and leg-banding analyses. Conservation Biology 16:789-799. Jones, M. E., D. Paetkau, E. Geffen, and C. Moritz. 2004. Genetic diversity and population structure of Tasmanian devils, the largest marsupial carnivore. Molecular Ecology 13:2197-2209. Julien-Laferriere, D., and M. Atramentowicz. 1990. Feeding and reproduction of three didelphid marsupials in two neotropical forests (French Guiana). Biotropica 22:404-415. Kaufman, D. W., and G. A. Kaufman. 1987. Reproduction by Peromyscus polionotus: number, size, and survival of offspring. Journal of Mammalogy 68:275-280. Kaufman, J. H. 1974. Social ethology of the whiptail wallaby. Animal Behaviour 22:281-369. Kaufmann, J. H. 1974. The ecology and evolution of social organization in the kangaroo family (Macropodidae). American Zoologist 14:51-62. Kayser, M., M. Krawczak, L. Excoffier, P. Dieltjes, D. Corach, V. Pascali, C. Gehrig, L. F. Bernini, J. Jespersen, E. Bakker, L. Roewer, and P. de Knijff. 2001. An extensive analysis of Y-chromosomal microsatellite haplotypes in globally dispersed human populations. The American Journal of Human Genetics 68:990-1018. Keller, L. F., and D. M. Waller. 2002. Inbreeding effects in wild populations. Trends in Ecology and Evolution 17:230-241. Kenagy, G. J., and S. C. Trombulak. 1986. Size and function of mammalian testes in relation to body size. Journal of Mammalogy 67:1-22. Kinnear, J. E., N. R. Sumner, and M. L. Onus. 2002. The red fox in Australia - an exotic predator turned biocontrol agent. Biological Conservation 108:335-359. Klinkova, E., M. Heistermann, and J. K. Hodges. 2004. Social parameters and urinary testosterone level in male chimpanzees (Pan troglodytes). Hormones and Behaviour 46:474-481. Knapp, L. A. 2005. The ABCs of MHC. Evolutionary Anthropology 14:28-37.

References 199

Kotiaho, J. S. 1999. Estimating fitness: comparison of body condition indices revisited. Oikos 87:399-400. Kraaijeveld-Smit, F. J. L., S. J. Ward, and P. D. Temple-Smith. 2002. Multiple paternity in a field population of a small carnivorous marsupial, the agile antechinus, Antechinus agilis. Behavioural Ecology and Sociobiology 52:84- 91. Kraaijeveld-Smit, F. J. L., S. J. Ward, and P. D. Temple-Smith. 2003. Paternity success and the direction of sexual selection in a field population of a semelparous marsupial, Antechinus agilis. Molecular Ecology 12:475-484. Krebs, C. J., and G. R. Singleton. 1993. Indices of condition for small mammals. Australian Journal of Zoology 41:317-323. Kubota, H., K. Watanabe, Y. Kakehi, and S. Watanabe. 2008. An assessment of genetic diversity in wild and captive populations of endangered Japanese bitterling Tanakia tanago (Cyprinidae) using amplified fragment length polymorphism (AFLP) markers. Fisheries Science 74:494-502. Lacy, R. C. 1989. Analysis of founder representation in pedigrees: founder equivalents and founder genome equivalents. Zoo Biology 8:111-123. Lacy, R. C., and J. D. Ballou 2002. Population Management 2000 User’s Manual. Chicago Zoological Society, Brookfield, USA. Land, D. E., D. Shindle, M. Cunningham, M. Lotz, and B. Ferree. 2004. Florida Panther Genetic Restoration and Management: Annual Performance Report 2003 - 2004. Florida Fish and Widlife Conservation Commission, Florida, USA. Lande, R., and S. J. Arnold. 1983. The measurement of selection on correlated characters. Evolution 37:1210-1226. Lee, A. K., and A. Cockburn 1985. The Evolutionary Ecology of Marsupials. Cambridge University Press, Cambridge, UK. Lewontin, R. C., and J. L. Hubby. 1966. A molecular approach to the study of genic heterozygosity in natural populations. II Amount of variation and degree of heterozygosity in natural populations of Drosophila psuedoobscura. Genetics 54:595-609.

References 200

Li, Y.-C., A. B. Korol, T. Fahima, A. Beiles, and E. Nevo. 2002. Microsatellites: genomic distribution, putative functions and mutational mechanisms: a review. Molecular Ecology 11:2453-2465. Lindenmayer, D., and M. Burgman 2005. Practical Conservation Biology. CSIRO Publishing, Collingwood, Australia. Loebel, D. A., R. K. Nurthen, R. Frankham, D. A. Briscoe, and D. Craven. 1992. Modeling problems in conservation genetics using Drosophila populations: consequences of equalising founder representation. Zoo Biology 11:319-332. Ludbrook, J., and H. Dudley. 1998. Why permutation tests are superior to t and F tests in biomedical research. American Statistician 52:127-132. Luikart, G., and P. E. England. 1999. Statistical analysis of microsatellite data. Trends in Ecology and Evolution 14:253-256. Luque-Larena, J. J., P. López, and J. Gosálbez. 2004. Spacing behaviour and morphology predict promiscuous mating strategies in the rock-dwelling snow vole, Chionomys nivalis. Canadian Journal of Zoology 82:1051-1060. Macdonald, A. J., N. Sankovic, S. D. Sarre, N. N. Fitzsimmons, M. J. Wakefield, J. A. M. Graves, and K. R. Zenger. 2006. Y chromosome microsatellite markers identified from the tammar wallaby (Macropus eugenii) and their amplification in three other macropod species. Molecular Ecology Notes 6:1202-1204. Madsen, T., R. Shine, M. Olsson, and H. Wittzell. 1999. Conservation biology: restoration of an inbred adder population. Nature 402:34-35. Main, A. R. 1961. The occurance of Macropodidae on islands and its climatic and ecological implications. Journal of the Royal Society of Western Australia 44:84-89. Mallet, J. 2005. Hybridisation as an invasion of the genome. Trends in Ecology and Evolution 20:229-237. Manly, B. F. J. 2001. Randomisation, Bootstrap and Monte Carlo Methods in Biology. Chapman & Hall, London, UK. Marshall, T. C., J. Slate, L. E. B. Kruuk, and J. M. Pemberton. 1998. Statistical confidence likelihood-based paternity inference in natural populations. Molecular Ecology 7:639-655. Martin, P., and P. Bateson 1993. Measuring Behaviour: An Introductory Guide. Cambridge University Press., Cambridge, UK.

References 201

Marvan, R., J. M. G. Stevens, A. D. Roeder, I. Mazura, M. W. Bruford, and J. R. de Ruiter. 2006. Male dominance rank, mating and reproductive success in captive bonobos (Pan paniscus). Folia Primatologica 77:364-376. Mathews, F., D. Moro, R. Strachan, M. Gelling, and N. Buller. 2006. Health surveillance in wildlife reintroductions. Biological Conservation 131:338-347. Matocq, M. D. 2004. Reproductive success and effective population size in woodrats (Neotoma macrotis). Molecular Ecology 13:1635-1642. Mawson, P. R. 2004a. Captive breeding programs and their contribution to Western Shield: Western Shield review February 2003. Conservation Science Western Australia 5:122-130. Mawson, P. R. 2004b. Translocations and fauna reintroduction sites: Western Shield review February 2003. Conservation Science Western Australia 5:108-121. Mays, H. L., and G. E. Hill. 2004. Choosing mates: good genes versus genes that are a good fit. Trends in Ecology and Evolution 19:554-559. McClure, P. A. 1981. Sex-biased litter reduction in food-restricted wood rats (Neotoma floridana). Science 211:1058-1060. McElligott, A. G., M. P. Gammell, H. C. Harty, D. R. Paini, D. T. Murphy, J. T. Walsh, and T. J. Hayden. 2001. Sexual size dimorphism in fallow deer (Dama dama): do larger, heavier males gain greater mating success? Behavioural Ecology and Sociobiology 49:266-272. McKenzie, L. M., and D. W. Cooper. 1997. Hybridisation between tammar wallaby (Macropus eugenii) populations from Western and South Australia. Heredity 88:398-400. McKenzie, N. L., A. A. Burbidge, and A. Baynes. 2006. Australian Mammal Map Updates. Western Australian Department of Conservation and Land Management. Department of Conservation and Land Management, Western Australia. McKenzie, S., E. M. Deane, and L. Burnett. 2002. Haematology and serum biochemistry of the tammar wallaby, Macropus eugenii. Comparative Clinical Pathology 11:229-237. Meyer, D. J., and J. W. Harvey 1998. Veterinary Laboratory Medicine: Interpretation and Diagnosis. W. B Saunders Company, Sydney, Australia.

References 202

Millar, J. S., and G. J. Hickling. 1990. Fasting endurance and the evolution of mammalian body size. Functional Ecology 4:5-12. Miller, E. J., M. D. B. Eldridge, and C. A. Herbert. in press. Dominance and paternity in the tammar wallaby (Macropus eugenii) in G. Coulson, and M. D. B. Eldridge, editors. Macropods: The Biology of Kangaroos, Wallabies and Rat- Kangaroos. Surrey Beatty and Sons, Sydney, Australia. Mills, H. R., D. Moro, and P. B. S. Spencer. 2004. Conservation significance of island versus mainland populations: a case study of dibblers (Parantechinus apicalis) in Western Australia. Animal Conservation 7:387-395. Milner-Gulland, E. J., O. M. Bukreeva, T. Coulson, A. A. Lushchekina, M. V. Kholodova, A. B. Bekenov, and I. A. Grachev. 2003. Reproductive collapse in a harem breeding ungulate. Nature 422:135. Milner-Gulland, E. J., and R. H. Mace. 1991. The impact of the ivory trade on the elephant population of the trade, as assessed by data from the trade. Biological Conservation 55:215-229. Montgomery, M. E., J. D. Ballou, R. K. Nurthen, P. R. England, D. A. Briscoe, and R. Frankham. 1997. Minimising kinship in captive breeding programs. Zoo Biology 16:377-389. Moore, J., and R. Ali. 1984. Are dispersal and inbreeding avoidance related? Animal Behaviour 32:94-112. Morin, P. A., and O. A. Ryder. 1991. Founder contribution and pedigree inference in a captive breeding colony of lion-tailed macaques, using mitochondrial DNA and DNA fingerprint analyses. Zoo Biology 10:341-352. Moritz, C. 1994. Defining 'Evolutionary Significant Units' for conservation. Trends in Ecology and Evolution 9:373-375. Moritz, C. 1999. Conservation units and translocations: strategies for conserving evolutionary processes. Hereditas 130:217-228. Moritz, C., A. Heideman, E. Geffen, and P. McCrae. 1997. Genetic population structure of the Greater Bilby Macrotis lagotis, a marsupial in decline. Molecular Ecology 6:925-936. Moseby, K. E., and E. O'Donnell. 2003. Reintroduction of the greater bilby, Macrotis lagotis (Reid) (Marsupialia:Thylacomyidae), to northern South Australia:

References 203

survival, ecology and notes on reintroduction protocols. Wildlife Research 30:15-27. Moss, G. L., and D. B. Croft. 1999. Body condition of the red kangaroo (Macropus rufus) in arid Australia: the effect of environmental condition, sex and reproduction. Australian Journal of Ecology 24:97-109. Muller, M. N., and R. W. Wrangham. 2004. Dominance, aggression and testosterone in wild chimpanzees: a test of the 'challenge hypothesis'. Animal Behaviour 67:113-123. Myers, N., and A. H. Knoll. 2001. The biotic crisis and the future of evolution. Proceedings of the National Academy of Sciences 98:5389-5392. Myers, P., and L. L. Master. 1983. Reproduction by Peromyscus maniculatus: size and compromise. Journal of Mammalogy 64:1-18. Mysterud, A., T. Coulson, and N. C. Stenseth. 2002. The role of males in the dynamics of ungulate populations. Journal of Animal Ecology 71:907-915. Nave, C. D. 2002. Fertility control in the eastern grey kangaroo, Macropus giganteus, Ph.D. Thesis. Department of Zoology. University of Melbourne, Melbourne, Australia. Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University Press, New York, USA. Nei, M., T. Maruyama, and R. Chakraborty. 1975. The bottleneck effect and genetic variability in populations. Evolution 29:1-10. Neilson, J. D., R. I. Perry, P. Valerio, and K. G. Waiwood. 1986. Condition of Atlantic cod Gadus morhua larvae after the transition to exogenous feeding: morphometrics, bouyancy and predator avoidance. Marine Ecology Progress Series 32:229-235. Neuhaus, P. 2000. Weight comparisons and litter size manipulation in Columbian ground squirrels (Spermophilus columbianus) show evidence of costs of reproduction. Behavioural Ecology and Sociobiology 48:75-83. Neveua, H., T. Hafenb, E. Zimmermannb, and Y. Rumplerc. 1998. Comparison of the genetic diversity of wild and captive groups of Microcebus murinus using the random amplified polymorphic DNA method. Folia Primatologica 69:127-135.

References 204

Newsome, A. E. 1965. The distribution of red kangaroos, Megaleia rufa (Desmarest), about sources of persistent food and water in central Australia. Australian Journal of Zoology 13:289-299. Nielsen, R. K., C. Pertoldi, and V. Loeschcke. 2007. Genetic evaluation of the captive breeding program of the Persian wild ass. Journal of Zoology 272:349-357. Noss, R. F. 1990. Can we maintain biological and ecological integrity? Conservation Biology 4:241-243. Nunn, C. L. 1999. The number of males in primate social groups: a comparative test of the socioecological model. Behavioural Ecology and Sociobiology 46:1-13. Nunney, L. 1991. The influence of age structure and fecundity on effective population size. Proceedings of the Royal Society B: Biological Sciences 246:71-76. Nunney, L. 1993. The influence of mating system and overlapping generations on effective population size. Evolution 47:1329-1341. Nunney, L., and D. R. Elam. 1994. Estimating the effective population size of conserved populations. Conservation Biology 8:175-184. Nussey, D. H., J. Pemberton, A. Donald, and L. E. B. Kruuk. 2006. Genetic consequences of human management in an introduced island population of red deer (Cervus elaphus). Heredity 97:56-65. O'Neal, D. M., D. G. Reichard, K. Pavilis, and E. D. Ketterson. 2008. Experimentally- elevated testosterone, female parental care, and reproductive success in a songbird, the Dark-eyed Junco (Junco hyemalis). Hormones and Behaviour 54:571-578. Ohta, T., and M. Kimura. 1973. A model of mutation appropriate to estimate the number of electrophoretically detectable alleles in a finite population. Genetical Research 22:201-204. Oliver, A. 1986. Social organisation and dispersal in the red kangaroo, Ph.D thesis. Murdoch University, Perth, Western Australia. Olsson, M., T. Madsen, J. Nordby, E. Wapstra, B. Ujvari, and H. Wittsell. 2003. Major histocompatibility complex and mate choice in sand lizards. Proceedings of the Royal Society B: Biological Sciences 270:254-S256. Orell, P. 2004. Fauna monitoring and staff training: Western Shield review - February 2003. Conservation Science Western Australia 5:51-95.

References 205

Oyama, K., M. Nojima, M. Shojo, M. Fukushima, K. Anada, and F. Mukai. 2007. Effect of sire mating patterns on future genetic merit and inbreeding in a closed beef cattle population. Journal of Animal Breeding and Genetics 124:73-80. Parrott, M., S. Ward, and P. Temple-Smith. 2007. Olfactory cues, genetic relatedness and female mate choice in the agile antechinus (Antechinus agilis). Behavioural Ecology and Sociobiology 61:1075-1079. Parrott, M. L., S. J. Ward, and P. D. Temple-Smith. 2006. Genetic similarity, not male size, influences female mate choice in the agile antechinus (Antechinus agilis). Australian Journal of Zoology 54:319-323. Pavey, C. 2006. National Recovery Plan for the Greater Bilby Macrotis lagotis. Northern Territory Department of Natural Resources, Environment and the Arts. 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. Pelletier, F., and M. Festa-Bianchet. 2006. Sexual selection and social rank in bighorn rams. Animal Behaviour 71:649-655. Peltier, W. R. 2004. Global glacial isostasy and the surface of the ice-age Earth: the ICE-5G (VM2) model and GRACE. Annual Review of Earth and Planetary Sciences 32:111-149. Perret, M. 1992. Environmental and social determinants of sexual function in the male lesser mouse lemur (Microcebus murinus). Folia Primatologica 59:1-25. Phillips, P. C., and S. J. Arnold. 1989. Visualising multivariate selection. Evolution 43:1209-1222. Piertney, S. B., and M. K. Oliver. 2006. The evolutionary ecology of the major histocompatibility complex. Heredity 96:7-21. Piry, S., G. Luikart, and J. M. Cornuet. 1999. BOTTLENECK: A computer program for detecting recent reductions in the effective population size using allele frequency data. Journal of Heredity 90:502-503. Pollak, J. P., R. C. Lacy, and J. D. Ballou. 2002. Population Management 2000, version 1.163. Chicago Zoological Society, Brookfield, USA. Poole, W. E. 2002. Eastern Grey Kangaroo Pages 335-338 in R. Strahan, editor. The Mammals of Australia. Reed New Holland, Sydney, Australia.

References 206

Poole, W. E., and P. E. Pilton. 1964. Reproduction in the grey kangaroo, Macropus canguru, in captivity. Wildlife Research 9:218-234. Poole, W. E., J. T. Wood, and N. G. Simms. 1991. Distribution of the tammar, Macropus eugenii, and the relationships of populations as determined by cranial morphometrics. Wildlife Research 18:625-639. Posada, D., and K. A. Crandall. 1998. MODELTEST: testing the model of DNA substitution. . Bioinformatics 14:817-818. Possingham, H., P. Jarman, and A. Kearns. 2003. Independent review of Western Shield. Department of Conservation and Land Management. Preston, B. T., C. Strobeck, J. M. Pemberton, and K. Wilson. 2001. Dominant rams lose out by sperm depletion. Nature 409:681-682. Pritchard, J. K., M. Stephens, and P. Donnelly. 2000. Inference of population structure using multilocus genotype data. Genetics 155:945-959. Pusey, A. E. 1987. Sex-biased dispersal and inbreeding avoidance in birds and mammals Trends in Ecology and Evolution 2:295-299. Qvarnström, A., and E. Forsgren. 1998. Should females prefer dominant males? Trends in Ecology and Evolution 13:498-501. Ralls, K., and J. Ballou. 1986. Captive breeding programs for populations with a small number of founders. Trends in Ecology and Evolution 1:19-22. Ramsay, M. A., and I. Stirling. 1986. On the mating system of polar bears. Canadian Journal of Zoology 64:2142-2151. Ramstad, K. M., C. A. Woody, G. K. Sgae, and F. W. Allendorf. 2004. Founding events influence genetic population structure of sockeye salmon (Oncorhynchus nerka) in Lake Clark, Alaska. Molecular Ecology 13:277-290. Rankmore, B. R., A. D. Griffiths, J. C. Z. Woinarski, B. L. Ganambarr, R. Taylor, K. Brennan, K. Firestone, and M. Cardoso. 2008. Island translocation of the northern quoll Dasyurus hallucatus as a conservation response to the spread of the cane toad Chaunus (Bufo) marinus in the Northern Territory, Australia. Australian Government National Heritage Trust. Northern Territory Government. Rasmussen, H. B., J. B. A. Okello, G. Wittemyer, H. R. Siegismund, P. Arctander, F. Vollrath, and I. Douglas-Hamilton. 2008. Age- and tactic-related paternity success in male African elephants. Behavioural Ecology 19:9-15.

References 207

Rathbun, G. B., B. B. Hatfield, and T. G. Murphey. 2000. Status of translocated sea otters at San Nicolas Island, California The Southwestern Naturalist 45:322- 328. Raymond, M., and F. Rousset. 2003. Genepop 3.4., an updated version of Genepop V.1.2 (1995): population genetics software for exact tests and ecumenicism. Journal of Heredity 86:248-249. Reed, D. H., and R. Frankham. 2003. Correlation between fitness and genetic diversity. Conservation Biology 17:230-237. Renfree, M. B., T. P. Fletcher, D. R. Blanden, P. R. Lewis, G. Shaw, K. Gordon, R. V. Short, E. Parer-Cook, and D. Parer. 1989. Physiological and behavioural events around the time of birth in macropodid marsupials. Pages 323-337 in G. Grigg, P. Jarman, and I. Hume, editors. Kangaroos, Wallabies and Rat-Kangaroos. Surrey Beatty and Sons Pty Limited, Sydney, Australia. Reynolds, J. D. 1996. Animal breeding systems. Trends in Ecology and Evolution 11:68-72. Rhymer, J. M., and D. Simberloff. 1996. Extinction by hybridisation and introgression. Annual Review of Ecology and Systematics 27:83-109. Rice, W. R. 1989. Analysing tables for statistical tests. Evolution 43:223-225. Ricker, W. E. 1975. Computation and interpretation of biological statistics of fish populations. Bulletin of the Fisheries Research Board of Canada 191:1-382. Riney, T. 1955. Evaluating condition of free-ranging red deer (Cervus elaphus) with special reference to New Zealand. New Zealand Journal of Science 36:429- 463. Risch, T. S., G. R. Michener, and F. S. Dobson. 2007. Variation in litter size: a test of hypotheses in Richardson's ground squirrels. Ecology 88:306-314. Robinson, N. A., N. D. Murray, and W. B. Sherwin. 1993. VNTR loci reveal differentiation between and structure within populations of the eastern barred bandicoot Perameles gunnii. Molecular Ecology 2:195-207. Røed, K. H., Ø. Holand, M. E. Smith, H. Gjøstein, J. Kumpula, and M. Nieminen. 2002. Reproductive success in reindeer males in a herd with varying sex ratio. Molecular Ecology 11:1239-1243. Rudd, C. D. 1994. Sexual behaviour of male and female tammar wallabies (Macropus eugenii) at post-partum oestrus. Journal of Zoology 232:151-162.

References 208

Rudd, C. D., R. V. Short, G. Shaw, and M. B. Renfree. 1996. Testosterone control of male-type sexual behaviour in the tammar wallaby (Macropus eugenii). Hormones and Behaviour 30:446-454. Rudnick, J. A., and R. C. Lacy. 2008. The impact of assumptions about founder relationships on the effectiveness of captive breeding strategies. Conservation Genetics 9:1439-1450. Russell, E. M. 1970. Agonistic interactions in the red kangaroo (Megaleia rufa). Journal of Mammalogy 51:80-88. Ryder, O. A. 1986. Species conservation and systematics: the dilemma of subspecies. Trends in Ecology and Evolution 1:9-10. Sanchez-Guzman, J. M., A. Villegas, C. Corbacho, R. Moran, A. Marzal, and R. Real. 2004. Response of the haematocrit to body condition changes in Northern Bald Ibis Geronticus eremita. Comparative Biochemistry and Physiology - Part A: Molecular and Integrative Physiology 139:41-47. Say, L., D. Pontier, and E. Natoli. 2001. Influence of oestrus synchronisation on male reproductive success in the domestic cat (Felis catus L.). Proceedings of the Royal Society of London B 268:1049-1053. Schonewald-Cox, C. M., S. M. Chambers, B. MacBryde, and W. L. Thomas 1983. Genetics and Conservation. Benjamin/Cummings, Menlo Park, USA. Schuett, G. W. 1997. Body size and agonistic experience affect dominance and mating success in male copperheads. Animal Behaviour 54:213-224. Schulte-Hostedde, A. I., and J. S. Millar. 2002. 'Little chipmunk' syndrome? Male body size and dominance in captive yellow-pine chipmunks (Tamias amoenus). Ethology 108:127-137. Schulte-Hostedde, A. I., and J. S. Millar. 2004. Intraspecific variation of testis size and sperm length in the yellow-pine chipmunk (Tamias amoenus): implications for sperm competition and reproductive success. Behavioural Ecology and Sociobiology 55:272-277. Schulte-Hostedde, A. I., J. S. Millar, and H. L. Gibbs. 2004. Sexual selection and mating patterns in a mammal with female-biased sexual size dimorphism. Behavioural Ecology 15:351-356. Schulte-Hostedde, A. I., J. S. Millar, and G. J. Hickling. 2001a. Evaluating body condition in small mammals. Journal of Canadian Zoology 79:1021-1029.

References 209

Schulte-Hostedde, A. I., J. S. Millar, and G. J. Hickling. 2001b. Sexual dimorphism in body composition of small mammals. Canadian Journal of Zoology-Revue Canadienne De Zoologie 79:1016-1020. Schulte-Hostedde, A. I., J. S. Millar, and G. J. Hickling. 2005a. Condition dependence of testis size in small mammals. Evolutionary Ecology Research 7:143-149. Schulte-Hostedde, A. I., B. Zinner, J. S. Millar, and G. J. Hickling. 2005b. Restitution of mass-size residuals: validating body condition indices. Ecology 86:155-163. Schwagmeyer, P. L., and E. D. Ketterson. 1999. Breeding synchrony and EPF rates: the key to a can of worms? Trends in Ecology and Evolution 14:47-48. Schwartz, M. K., D. A. Tallmon, and G. Luikart. 1998. Review of DNA-based census and effective population size estimators. Animal Conservation 1:293-299. Seddon, N., W. Amos, A. M. Raoul, and J. A. Tobias. 2004. Male heterozygosity predicts territory size, song structure and reproductive success in a cooperatively breeding bird. Proceedings of the Royal Society of London B: Biological Sciences 271:1823-1829. Seddon, P. J., D. P. Armstrong, and R. F. Maloney. 2007. Developing the science of reintroduction biology. Conservation Biology 21:303-312. Setchell, J. M., and A. F. Dixson. 2001. Changes in the secondary sexual adornments of male mandrills (Mandrillus sphinx) are associated with gain and loss of alpha status. Hormones and Behaviour 39:177-184. Setchell, J. M., T. Smith, E. J. Wickings, and L. A. Knapp. 2008. Social correlates of testosterone and ornamentation in male mandrills. Hormones and Behaviour 54:365-372. Seymour, A. M., M. E. Montgomery, B. H. Costello, S. Ihle, G. Johnsson, B. St John, D. Taggart, and B. A. Houlden. 2001. High effective inbreeding coefficients correlate with morphological abnormalities in populations of South Australian koalas (Phascolarctos cinereus). Animal Conservation 4:211-219. Shaffer, M. 1981. Minimum population sizes for species conservation. BioScience 31:131-134. Shepherd, N. 1987. Condition and recruitment of kangaroos. Pages 135-158 in G. Caughley, N. Shepherd, and J. Short, editors. Kangaroos: Their Ecology and Management in the Sheep Rangelands in Australia. Cambridge University Press, Cambridge, UK.

References 210

Sherwin, W. B., P. Timms, J. Wilcken, and B. Houlden. 2000. Analysis and conservation implications of koala genetics. Conservation Biology. 14: 639- 649. Short, J., and A. Smith. 1994. Mammal decline and recovery in Australia. Journal of Mammalogy 75:288-297. Short, R. V. 1979. Sexual selection and its component parts, somatic and genital selection, as illustrated by man and the great apes. Advances in the Study of Behaviour 9:131-158. Sigg, D. P. 2006. Reduced genetic diversity and significant genetic differentiation after translocation: comparison of the remnant and translocated populations of bridled nailtail wallabies (Onychogalea fraenata). Conservation Genetics 7:577-589. Simon, A., D. Thomas, J. Blondel, P. Perret, and M. M. Lambrechts. 2004. Physiological ecology of Mediterranean Blue Tits (Parus caeruleus L.): effects of ectoparasites (Protocalliphora spp.) and food abundance on metabolic capacity of nestlings. Physiological and Biochemical Zoology 77:492-501. Sinclair, E. A. 2001. Phylogeographic variation in the quokka, Setonix brachyurus (Marsupialia: Macropodidae): implications for conservation. Animal Conservation 4:325-333. Slatkin, M. 1995. A measure of population subdivision based on microsatellite allele frequencies. Genetics 139:457-462. Smith, M. J., and L. Hinds. 2002. Tammar Wallaby. Pages 329-331 in R. Strahan, editor. The Mammals of Australia. Reed New Holland, Sydney, Australia. Solberg, E. J., A. Loison, T. H. Ringsby, B. E. Saether, and M. Heim. 2002. Biased adult sex ratio can affect fecundity in primiparous moose Alces alces. Wildlife Biology 8:117-128. Soulé, M. E. 1976. Allozyme variation, its determinants in space and time. Pages 151- 169 in F. J. Ayala, editor. Molecular Evolution. Sinauer Associates Massachusetts, USA. Soulé, M. E. 1985. What is conservation biology? BioScience 35:727-734. Soulé, M. E., M. Gilpin, W. Conway, and T. Foose. 1986. The millennium ark: how long a voyage, how many staterooms, how many passengers? Zoo Biology 5:101-113.

References 211

Soulé, M. E., and B. M. Wilcox 1980. Conservation Biology: An Evolutionary- Ecological Perspective. Sinauer, Sunderland, USA. Southgate, R. 2005. Age classes of the greater bilby (Macrotis lagotis) based on track and faecal pellet size. Wildlife Research 32:625-630. Southgate, R. I. 1990. Distribution and abundance of the greater bilby, Macrotis lagotis Reid (Marsupialia: Peramelidae). Pages 293-302 in J. H. Seebeck, P. R. Brown, R. I. Wallis, and C. M. Kemper, editors. Bandicoots and Bilbies. Surrey Beatty and Sons, Sydney, Australia. Southgate, R. I., and M. Adams. 1993. Genetic variation in the greater bilby (Macrotis lagotis). Pacific Conservation Biology 1:46-52. Southgate, R. I., P. Christie, and K. Bellchambers. 2000. Breeding biology of captive, reintroduced and wild greater bilbies, Macrotis lagotis (Marsupialia:Peramelidae). Wildlife Research 27:621-628. Spencer, P. B. S., D. M. Odorico, J. Jones, H. D. Marsh, and D. J. Miller. 1995. Highly variable microsatellites in isolated colonies of the rock-wallaby (Petrogale assimilis). Molecular Ecology 4:523-525. Spielman, D., and R. Frankham. 1992. Modeling problems in conservation genetics using captive Drosophila populations: improvement of reproductive fitness due to immigration of one individuals into small partially inbred populations. Zoo Biology 11:343-351. Spinage, C. A. 1984. Seasonal influences and the kidney fat index in two equatorial African ungulates. African Journal of Ecology 22:217-221. Spong, G. F., S. J. Hodge, A. J. Young, and T. H. Clutton-Brock. 2008. Factors affecting the reproductive success of dominant male meerkats. Molecular Ecology 17:2287-2299. Stinchcombe, J. R., A. F. Agrawal, P. Hohenlohe, S. J. Arnold, and M. W. Blows. 2008. Estimating nonlinear selection gradients using quadratic regression coefficients: double or nothing? Evolution 62:2435-2440. Stirrat, S. C. 2003. Body condition and blood chemistry in agile wallabies (Macropus agilis) in the wet-dry tropics. Wildlife Research 30:59-67. Storr, G. M. 1960. The physiography, vegetation and vertebrate fauna of North Island, Houtman Abrolhos. Journal of the Royal Society of Western Australia 43:59- 62.

References 212

Storr, G. M. 1965. The physiography, vegetation and vertebrate fauna of the Wallabi Group, Houtman Abrolhos. Journal of the Royal Society of Western Australia 48:1-14. Strahan, R. 2002. The Mammals of Australia. Reed New Holland, Sydney, Australia. Sunnucks, P. 2000. Efficient genetic markers for population biology. Trends in Ecology and Evolution 15:199-203. Sunnucks, P., and D. Hales. 1996. Numerous transposed sequences of mitochondrial cytochrome oxidase I-II in aphids of the genus Sitobion (Hemiptera: Aphidae). Molecular Ecology 13:510-524. Svendsen, G. 1964. Comparative reproduction and development in two species of mice in the genus Peromyscus. Transactions of the Kansas Academy of Science (1903-) 67:527-538. Swofford, D. L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony (*and other methods). Version 4. Sinauer Associates, Sunderland, USA. Takatsuki, S. 2000. Kidney fat and marrow fat indices of the sika deer population at Mount Goyo, northern Japan. Ecological Resarch 15:453-457. Tallmon, D. A., A. Koyuk, G. Luikart, and M. A. Beaumont. 2008. ONeSAMP: a program to estimate effective population size using approximate Bayesian computation. Molecular Ecology Resources 8:299-301. Tallmon, D. A., G. Luikart, and R. S. Waples. 2004. The alluring simplicity and complex reality of genetic rescue. Trends in Ecology and Evolution 19:489- 496. Tarr, C. L., S. Conant, and R. C. Fleisher. 1998. Founder events and variation at microsatellite loci in an insular passerine bird, the Laysan finch (Telespiza cantans). Molecular Ecology 7:719-731. Taylor, A. C., and D. W. Cooper. 1998. A set of tammar wallaby (Macropus eugenii) microsatellites tested for genetic linkage. Molecular Ecology 7:925-931. Taylor, A. C., and D. W. Cooper. 1999. Microsatellites identify introduced New Zealand tammar wallabies (Macropus eugenii) as an 'extinct' taxon. Animal Conservation 2:41-49. Temple, S. A. 1990. The nasty necessity: eradicating exotics. Conservation Biology 5:113-115.

References 213

Tenhumberg, B., A. J. Tyre, A. R. Pople, and H. Possingham. 2004. Do harvest refuges buffer kangaroos against evolutionary responses to selective harvesting? Ecology 85:2003-2017. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 25:4876- 4882. Tregenza, T., and N. Wedell. 2000. Genetic compatibility, mate choice and patterns of parentage: invited Review. Molecular Ecology 9:1013-1027. Trivers, R. L. 1972. Parental investment and sexual selection. Pages 136-172 in B. Campbell, editor. Sexual Selection and the Descent of Man, 1871-1971. Aldine-Atherton, Chicago, USA. Tsvey, A., V. Bulyuk, and V. Kosarev. 2007. Influence of body condition and weather on departures of first-year European robins, Erithacus rubecula, from an autumn migratory stopover site. Behavioural Ecology and Sociobiology 61:1665-1674. Tudge, C. 1995. Captive audiences for future conservation. New Scientist 145:51-52. Tyndale-Biscoe, C. H., and M. B. Renfree 1987. Reproductive Physiology of Marsupials. Cambridge University Press, Cambridge, UK. van Oosterhout, C., W. F. Hutchinson, D. P. M. Wills, and P. Shipley. 2004. MICRO- CHECKER: software for identifying and correcting genotyping errors in microsatellite data. . Molecular Ecology Notes 4:535-538. van Rooyen, A. F. 1993. Variation in body condition of impala and nyala in relation to social status and reproduction. South African Journal of Wildlife Research 23:36-38. Vervust, B., S. P. Lailvaux, I. Grbac, and R. Van Damme. 2008. Do morphological condition indices predict locomotor performance in the lizard Podarcis sicula? Acta Oecologica 34:244-251. Vicente, J., L. Pérez-Rodríguez, and C. Gortazar. 2007. Sex, age, spleen size, and kidney fat of red deer relative to infection intensities of the lungworm Elaphostrongylus cervi. Naturwissenschaften 94:581-587. Vié, J.-C., C. Hilton-Taylor, C. Pollock, J. Ragle, J. Smart, S. N. Stuart, and R. Tong 2008. The IUCN Red List: a key conservation tool. IUCN Gland, Switzerland.

References 214

Viggers, K. L., and D. B. Lindemayer. 2001. Haematological and plasma biochemical values of the greater glider in Australia. Journal of Wildlife Diseases 37:370- 374. Viggers, K. L., D. B. Lindemayer, R. B. Cunningham, and C. F. Donnelly. 1998. Estimating body condition in the mountain brushtail possum, Trichosurus caninus. Wildlife Research 25:499-509. Vilà, C., A. Sundqvist, Ø. Flagstad, J. Seddon, S. Björnerfeldt, I. Kojola, A. Casulli, S. H., P. Wabakken, and H. Ellegren. 2003. Rescue of a severely bottlenecked wolf (Canis lupus) population by a single immigrant. Proceedings of the Royal Society B: Biological Sciences 270:91-97. Villegas, A., J. M. Sánchez, E. Costillo, and C. Corbacho. 2002. Blood chemistry and hematocrit of the black vulture (Aegypius monachus). Comparative Biochemistry and Physiology - Part A: Molecular and Integrative Physiology 132:489-497. Vogelnest, L., and T. Portas. 2008. Macropods. Pages 133-225 in L. Vogelnest, and R. Woods, editors. Medicine of Australian Mammals. CSIRO Publishing, Victoria, Australia. von Schantz, T., G. Goransson, G. Andersson, I. Froberg, M. Grahn, A. Helgee, and H. Wittzell. 1989. Female choice selects for a viability-based male trait in pheasants. Nature 337:166-169. von Schantz, T., H. Wittzell, G. G., and M. Grahn. 1997. Mate choice, male condition- dependent ornamentation and MHC in the pheasant. Hereditas 127:133-140. von Schantz, T., H. Wittzell, G. Goransson, M. Grahn, and K. Persson. 1996. MHC genotype and male ornamentation: genetic evidence for the Hamilton-Zuk model. Proceedings of the Royal Society of London B: Biological Sciences 263:265-271. Wang, J. L., and N. Ryman. 2001. Genetic effects of multiple generations of supportive breeding. Conservation Biology 15:1619-1631. Waples, R. S. 1991. Pacific salmon, oncorhynchus spp., and the definition of "species" under the endangered species act Marine Fisheries Review 53:11. Weatherhead, P. J., and G. P. Brown. 1996. Measurement versus estimation of body condition in snakes. Canadian Journal of Zoology 74:1617-1621.

References 215

Weatherhead, P. J., M. R. Prosser, H. L. Gibbs, and G. P. Brown. 2002. Male reproductive success and sexual selection in northern water snakes determined by microsatellite DNA analysis. Behavioural Ecology 13:808-815. Weber, J. L., and C. Wong. 1993. Mutation of human short tandem repeats. Human Molecular Genetics 2:1123-1128. Weir, B. S., and C. C. Cockerham. 1984. Estimating F-statistics for the analysis of population structure. Evolution 38:1358-1370. Westneat, D. F., and P. W. Sherman. 1997. Density and extra-pair fertilisations in birds: a comparative analysis. Behavioural Ecology and Sociobiology 41:205- 215. Whitfield, D. P., W. Cresswell, N. P. Ashmole, N. A. Clark, and A. D. Evans. 1999. No evidence for sparrowhawks selecting redshanks according to size or condition. Journal of Avian Biology 30:31-39. Whitlock, M. C., P. K. Ingvarsson, and T. Hatfield. 2000. Local drift load and the heterosis of interconnected populations. Heredity 84:452-457. Wickings, E. J., and A. F. Dixson. 1992. Testicular function, secondary sexual development, and social-status in male mandrills (Mandrillus sphinx). Physiology and Behaviour 52:909-916. Wicks, R. M., and P. Clark. 2005. Clinical haematology of the southern brown bandicoot (Isoodon obesulus). Comparative Clinical Pathology 14:56-60. Williamson-Natesan, E. 2005. Comparison of methods for detecting bottlenecks from microsatellite loci. Conservation Genetics 6:551-562. Williamson, P., T. P. Fletcher, and M. B. Renfree. 1990. Testicular development and maturation of the hypothalamic-pituitary-testicular axis in the male tammar, Macropus eugenii. Journal of Reproduction and Fertility 88:549-557. Wingfield, J. C., R. E. Hegner, A. M. Dufty, Jr., and G. F. Ball. 1990. The "Challenge Hypothesis": theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. American Naturalist 136:829-846. Wisely, S. M., D. B. McDonald, and S. W. Buskirk. 2003. Evaluation of the genetic management of the endangered black-footed ferret (Mustela nigripes). Zoo Biology 22:287-298.

References 216

Wittenberger, J. F. 1979. The evolution of mating systems in birds and mammals. Pages 271-349 in P. Marler, and J. G. Vandenbergh, editors. Handbook of Behavioural Neurobiology. Plenum, New York, USA. Wittenberger, J. F. 1981. Animal Social Behaviour. Duxbury Press, Boston, USA. Worthington Wilmer, J., P. J. Allen, P. P. Pomeroy, S. D. Twiss, and W. Amos. 1999. Where have all the fathers gone? An extensive microsatellite analysis of paternity in the grey seal (Halichoerus grypus). Molecular Ecology 8:1417- 1429. Wright, M., and P. Stott. 1999. The Kangaroo Island Tammar wallaby: assessing ecologically sustainable commercial harvesting. A report for the Rural Industries Research and Development Corporation. University of Adelaide, South Australia, Canberra, Australia. Wright, S. 1931. Evolution in Mendelian populations. Genetics 16:97-159. Wright, S. 1940. Breeding structure of populations in relation to speciation. American Naturalist 74:232-248. Wright, S. 1969. Evolution and the Genetics of Populations: The Theory of Gene Frequencies. University of Chicago Press, Chicago, USA. Young, A., T. Boyle, and T. Brown. 1996. The population genetic consequences of habitat fragmentation for plants. Trends in Ecology and Evolution 11:413-418. Zahavi, A. 1975. Mate selection - a selection for a handicap. Journal of Theoretical Biology 67:205-214. Zenger, K. R., and D. W. Cooper. 2001. Characterisation of 14 macropod microsatellite genetic markers. Animal Genetics 32:160-167. Zenger, K. R., M. D. B. Eldridge, L. C. Pope, and D. W. Cooper. 2003. Characterisation and cross-species utility of microsatellite markers within kangaroos, wallabies and rat kangaroos (Macropodoidea: Marsupialia). Australian Journal of Zoology 51:587-596. Zink, R. M., and G. F. Barrowclough. 2008. Mitochondrial DNA under seige in avian phylogeography. Molecular Ecology 17:2107-2121.

Appendix 1 217

Appendix 1

DNA extraction protocol

DNA extraction was performed using a salting out method (Sunnucks & Hales 1996). Briefly, each tissue sample was placed in a 1.8ml microcentrifuge tube with 15l proteinase K (10mg/ml) and 580l TNES (50mM Tris base pH 7.5, 400mM NaCl, 20mM EDTA, 0.5% SDS) and placed in a water bath at 50ºC to digest overnight. Following tissue digestion, 170l 5M NaCl was added to each sample, vortexed for 15 s, and then centrifuged at 14000 rpm for 10 min. The samples were then removed from the centrifuge and 600 – 640l of the supernatant was removed and transferred into a new tube. 600l of 100% ethanol (at -20ºC) was then added to each tube and inverted gently to aggregate the DNA. The samples were then centrifuged at 14 000 rpm for five min to pellet the DNA with excess ethanol then being poured off. The DNA pellet was then washed with 300l of 70% ethanol and again centrifuged at 14 000 rpm for five min. Excess ethanol was then discarded and the DNA pellet air- dried. The pellet was then reconstituted with 150l of milli-Q H2O at 65ºC for 30 min until the pellet was dissolved. The samples were diluted 1:25 (DNA: milli-Q H2O) and placed on a mechanical rotator for 24 to 48 hours to ensure the diluted DNA samples were mixed thoroughly. The amount of DNA extracted for each sample was quantified using a NanoDrop® ND-1000 Spectrophotometer (BioLab, New Zealand).

Appendix 2 218

Appendix 2

Allele frequencies for seven microsatellite loci for the Return to Dryandra (Western Australia) captive breeding colony for the greater bilby (Macrotis lagotis) between 1998 and 2005. Alleles for locus Bil02 Population 168 170 172 174 176 178 180 98/99 0.130 0.283 0.196 0.065 0.174 0.152 - 2000 0.088 0.284 0.225 0.098 0.186 0.118 - 2001 0.113 0.238 0.338 0.075 0.138 0.100 - 2002 0.173 0.173 0.318 0.109 0.145 0.082 - 2003 0.173 0.214 0.268 0.071 0.161 0.107 0.006 2004 0.221 0.188 0.286 0.071 0.156 0.071 0.006 2005 0.173 0.253 0.273 0.067 0.160 0.060 0.013

Alleles for locus Bil16 206 208 210 212 98/99 0.813 0.083 0.042 0.063 2000 0.721 0.183 - 0.096 2001 0.775 0.125 - 0.100 2002 0.714 0.080 0.080 0.125 2003 0.696 0.071 0.155 0.077 2004 0.701 0.058 0.156 0.084 2005 0.633 0.060 0.220 0.087

Alleles for locus Bil22 183 189 191 193 195 197 98/99 0.063 0.125 0.208 0.438 - 0.167 2000 0.087 0.163 0.183 0.481 - 0.087 2001 0.063 0.100 0.275 0.450 - 0.113 2002 0.080 0.054 0.152 0.518 - 0.196 2003 0.137 0.060 0.185 0.500 - 0.119 2004 0.169 0.045 0.266 0.429 - 0.091 2005 0.200 0.060 0.253 0.427 0.007 0.053

Appendix 2 219

Alleles for locus Bil41 242 244 246 248 250 252 256 260 98/99 0.021 - 0.167 0.063 0.021 0.500 0.188 0.042 2000 - - 0.212 0.038 0.019 0.452 0.279 - 2001 - 0.025 0.238 0.213 0.025 0.338 0.150 0.013 2002 - 0.009 0.223 0.304 0.036 0.295 0.107 0.027 2003 - 0.006 0.226 0.268 0.024 0.304 0.125 0.048 2004 - - 0.253 0.253 0.045 0.318 0.091 0.039 2005 - - 0.193 0.220 0.053 0.353 0.133 0.047

Alleles for locus Bil55 167 181 185 187 189 191 193 195 197 98/99 0.063 - 0.208 0.063 0.083 0.396 0.021 - 0.167 2000 0.038 - 0.154 0.019 0.154 0.500 0.029 0.010 0.096 2001 0.038 - 0.325 0.025 0.075 0.388 0.013 - 0.138 2002 0.071 - 0.277 0.027 0.116 0.366 0.027 0.009 0.107 2003 0.125 - 0.208 0.030 0.137 0.369 0.042 0.012 0.077 2004 0.123 - 0.188 0.013 0.149 0.351 0.052 0.006 0.117 2005 0.087 0.007 0.147 0.013 0.247 0.347 0.040 - 0.113

Alleles for locus Bil56 229 233 235 239 241 243 98/99 0.375 0.188 0.104 0.229 0.104 - 2000 0.529 0.144 0.087 0.115 0.125 - 2001 0.564 0.077 0.128 0.128 0.090 0.013 2002 0.509 0.018 0.098 0.214 0.152 0.009 2003 0.524 0.065 0.113 0.173 0.107 0.018 2004 0.500 0.058 0.162 0.175 0.078 0.026 2005 0.440 0.080 0.240 0.160 0.047 0.033

Alleles for locus Bil63 229 231 233 235 237 98/99 0.292 - 0.313 0.104 0.292 2000 0.221 0.019 0.192 0.106 0.462 2001 0.350 - 0.150 0.175 0.325 2002 0.313 - 0.277 0.161 0.250 2003 0.220 - 0.321 0.143 0.315 2004 0.188 - 0.312 0.175 0.325 005 0.153 - 0.300 0.133 0.413 Appendix 3 220

Appendix 3

Allele frequencies for seven microsatellite loci for the Peron Captive Breeding Centre (Western Australia) captive breeding colony for the greater bilby (Macrotis lagotis) between 1997 and 2007.

Alleles for locus Bil02 Population 162 168 170 172 174 176 178 180 97/98 0.091 - 0.545 0.045 - 0.273 - 0.045 1999 0.091 0.023 0.409 0.148 0.045 0.182 0.023 0.080 2000 0.089 0.009 0.357 0.179 0.054 0.152 0.054 0.107 2001 0.039 0.049 0.431 0.137 0.029 0.196 0.029 0.088 2002 0.056 0.120 0.345 0.042 0.141 0.141 0.070 0.085 2003 0.042 0.181 0.211 0.036 0.241 0.072 0.157 0.060 2004 0.038 0.217 0.170 0.019 0.302 0.085 0.151 0.019 2005 0.069 0.167 0.137 0.010 0.304 0.078 0.176 0.059 2006 0.143 0.083 0.214 0.024 0.238 0.024 0.179 0.095 2007 0.178 0.033 0.233 0.022 0.244 0.033 0.189 0.067

Alleles for locus Bil16 204 206 208 210 212 97/98 0.227 0.545 0.182 0.045 - 1999 0.080 0.614 0.216 0.045 0.045 2000 0.071 0.563 0.232 0.080 0.054 2001 0.147 0.539 0.255 0.039 0.020 2002 0.085 0.577 0.232 0.077 0.028 2003 0.030 0.596 0.217 0.114 0.042 2004 0.038 0.547 0.198 0.142 0.075 2005 0.029 0.549 0.235 0.118 0.069 2006 0.071 0.500 0.250 0.083 0.095 2007 0.078 0.478 0.233 0.100 0.111

Appendix 3 221

Alleles for locus Bil22 183 185 187 189 191 193 195 197 97/98 - 0.045 - 0.136 0.591 0.091 0.136 - 1999 0.034 0.068 0.080 0.114 0.386 0.182 0.114 0.023 2000 0.027 0.063 0.080 0.107 0.384 0.232 0.098 0.009 2001 0.078 0.029 0.059 0.088 0.412 0.137 0.147 0.049 2002 0.063 0.049 0.035 0.049 0.444 0.085 0.169 0.106 2003 0.060 0.060 0.024 0.036 0.470 0.114 0.120 0.114 2004 0.047 0.094 0.009 0.075 0.377 0.142 0.123 0.132 2005 0.029 0.078 0.029 0.069 0.382 0.108 0.186 0.118 2006 - 0.071 0.048 0.095 0.262 0.167 0.262 0.095 2007 - 0.067 0.044 0.122 0.289 0.211 0.211 0.056

Alleles for locus Bil41 242 244 246 248 250 252 256 260 97/98 0.091 0.045 0.455 0.091 0.091 0.045 0.045 0.136 1999 0.080 0.057 0.364 0.057 0.136 0.182 0.034 0.091 2000 0.063 0.036 0.330 0.071 0.152 0.241 0.036 0.071 2001 0.049 0.010 0.422 0.049 0.127 0.176 0.069 0.098 2002 0.063 0.007 0.507 0.035 0.070 0.169 0.035 0.113 2003 0.048 0.012 0.548 0.078 0.054 0.187 0.012 0.060 2004 0.085 0.047 0.396 0.094 0.038 0.226 0.028 0.085 2005 0.078 - 0.422 0.098 0.078 0.196 0.020 0.108 2006 0.119 - 0.310 0.119 0.095 0.214 0.036 0.107 2007 0.100 - 0.233 0.167 0.089 0.233 0.056 0.122

Alleles for locus Bil55 167 181 185 187 189 191 193 195 197 97/98 0.182 - 0.091 0.136 0.364 0.091 - 0.136 - 1999 0.136 - 0.136 0.068 0.386 0.102 - 0.159 0.011 2000 0.098 - 0.125 0.071 0.464 0.116 - 0.125 - 2001 0.206 - 0.157 0.088 0.382 0.069 - 0.088 0.010 2002 0.246 - 0.190 0.099 0.275 0.063 0.007 0.092 0.028 2003 0.217 - 0.199 0.084 0.313 0.048 0.012 0.078 0.048 2004 0.075 0.019 0.132 0.142 0.396 0.047 0.009 0.113 0.066 2005 0.118 0.029 0.137 0.127 0.343 0.029 0.010 0.157 0.049 2006 0.036 0.143 0.119 0.083 0.429 0.048 - 0.107 0.036 2007 0.011 0.178 0.100 0.078 0.400 0.067 - 0.144 0.022 Appendix 3 222

Alleles for locus Bil56 229 233 235 239 243 245 247 249 251 97/98 - 0.091 - - 0.136 0.091 0.545 0.045 0.091 1999 0.011 0.034 0.023 0.034 0.136 0.057 0.432 0.023 0.250 2000 0.009 0.009 0.018 0.125 0.170 0.036 0.393 0.018 0.223 2001 0.010 - 0.010 0.127 0.137 0.029 0.441 0.029 0.216 2002 0.014 0.007 0.014 0.007 0.155 0.296 0.268 0.056 0.183 2003 - 0.006 0.018 - 0.108 0.452 0.163 0.072 0.181 2004 0.019 0.028 0.009 0.038 0.094 0.472 0.104 0.028 0.208 2005 0.039 0.020 0.029 0.049 0.108 0.353 0.127 0.078 0.196 2006 0.085 0.024 0.110 0.037 0.146 0.232 0.085 0.085 0.195 2007 0.080 0.034 0.136 0.057 0.148 0.159 0.091 0.080 0.216

Alleles for locus Bil63 229 231 233 235 237 241 97/98 - 0.045 0.045 0.773 0.091 0.045 1999 - 0.068 0.057 0.773 0.057 0.045 2000 - 0.054 0.045 0.813 0.045 0.045 2001 - 0.118 0.020 0.755 0.049 0.059 2002 0.014 0.243 0.050 0.586 0.050 0.057 2003 0.012 0.293 0.030 0.585 0.037 0.043 2004 0.047 0.264 0.160 0.491 0.019 0.019 2005 0.170 0.180 0.300 0.240 0.050 0.060 2006 0.119 0.119 0.440 0.250 0.060 0.012 2007 0.122 0.100 0.456 0.244 0.067 0.011 Appendix 4 223

Appendix 4

Overall allele frequencies at seven microsatellite loci for two Western Australian captive bred populations of greater bilbies (Macrotis lagotis). Dryandra =Return to Dryandra; Peron = Peron captive Breeding Centre.

Alleles at locus Bil02 Population 162 168 170 172 174 176 178 180 Dryandra - 0.151 0.237 0.237 0.077 0.188 0.100 0.009 Peron 0.090 0.115 0.244 0.058 0.192 0.102 0.122 0.077

Alleles at locus Bil16 204 206 208 210 212 Dryandra - 0.697 0.0990.115 0.089 Peron 0.060 0.551 0.2220.113 0.055

Alleles at locus Bil22 183 185 187 189 191 193 195 197 Dryandra 0.135 - - 0.089 0.220 0.456 0.005 0.094 Peron 0.047 0.060 0.028 0.077 0.402 0.158 0.141 0.086

Alleles at locus Bil41 242 244 246 248 250 252 256 260 Dryandra 0.007 0.007 0.211 0.177 0.039 0.356 0.167 0.037 Peron 0.066 0.023 0.406 0.096 0.088 0.201 0.032 0.088

Alleles at locus Bil55 167 181 185 187 189 191 193 195 197 Dryandra 0.101 0.005 0.190 0.021 0.158 0.385 0.037 0.007 0.096 Peron 0.141 0.039 0.152 0.096 0.352 0.070 0.006 0.115 0.030

Appendix 4 224

Alleles at locus Bil56 229 233 235 239 241 243 245 247 249 251 Dryandra 0.484 0.099 0.143 0.150 0.104 0.021 - - - - Peron 0.023 0.017 0.034 0.053 - 0.140 0.273 0.199 0.053 0.208

Alleles at locus Bil63 229 231 233 235 237 241 Dryandra 0.206 0.005 0.289 0.135 0.365 - Peron 0.057 0.180 0.1590.511 0.051 0.042

Appendix 5 225

Appendix 5

Allele frequencies for nine autosomal microsatellite loci for the Houtman Abrolhos Archipelago tammar wallaby (Macropus eugenii) populations from East Wallabi Island (EWI), West Wallabi Island (WWI) and North Island (NI), Western Australia.

Alleles for locus G31-1 Population 117 119 121 131 133 EWI 0.014 0.957 0.029 - - WWI 0.567 - - 0.067 0.367 NI 0.583 - - - 0.417

Alleles for locus T46-5 158 162 166 170 174 178 182 EWI - - 0.132 0.191 0.397 0.250 0.029 WWI 0.033 0.183 0.133 0.350 0.167 0.133 - NI - 0.056 0.542 0.403 - - -

Alleles for locus T31-1 106 108 112 114 116 EWI 0.057 0.657 0.214 0.071 - WWI - - 0.050 0.033 0.917 NI - - 0.417 - 0.583

Alleles for locus G16-1 EWI 154 158 160 WWI - 0.957 0.043 NI 0.914 0.086 - EWI 0.985 - 0.015

Alleles for locus T3-1T 226 230 234 238 242 246 250 294 EWI 0.029 0.129 0.243 0.114 0.343 0.086 0.057 - WWI ------0.017 NI ------0.139

Appendix 5 226

Alleles for locus G26-4 177 181 EWI 0.986 0.014 WWI 0.650 0.350 NI 0.736 0.264

Alleles for locus Me2 247 249 251 257 259 261 263 EWI 0.471 0.397 0.118 - 0.015 - - WWI - - - 0.200 0.383 0.217 0.200 NI - - - 0.014 0.931 0.042 0.014

Alleles for locus Me14 162 164 166 168 170 172 EWI 0.757 0.014 0.186 0.043 - - WWI - 0.117 0.283 0.467 0.017 0.117 NI - 0.042 0.639 0.319 - -

Alleles for locus G20-2 133 139 143 145 151 EWI 0.914 0.086 - - - WWI 0.054 - 0.107 0.821 0.018 NI - - - 1.000 -

Appendix 6 227

Appendix 6

Allele frequencies for four Y-linked microsatellite loci for the Houtman Abrolhos Archipelago tammar wallaby (Macropus eugenii) populations from East Wallabi Island (EWI), West Wallabi Island (WWI) and North Island (NI), Western Australia.

Alleles for locus MeY01 Population 310 314 EWI 1.000 - WWI - 1.000 NI - 1.000

Alleles for locus MeY28 325 329 331 EWI 1.000 - - WWI - 0.188 - NI - 0.813 1.000

Alleles for locus MeY37A 151 EWI 1.000 WWI 1.000 NI 1.000

Alleles for locus MeY37B 169 EWI 1.000 WWI 1.000 NI 1.000

Appendix 7 228

Appendix 7

Summary of mtDNA control region (642 bp) variable sites and frequency of each haplotype for the Houtman Abrolhos Archipelago tammar wallaby (Macropus eugenii) populations from East Wallabi Island (EWI), West Wallabi Island (WWI) and North Island (NI), Western Australia.

Variable sites (base pair) Frequency Haplotype 285 314 349 361 385 400 409 421 467 EWI WWI NI A T T A T T T C C C 29 0 0 B C C A T T T C C C 3 0 0 D T T A T T C C C C 1 0 0 E T T A C C T C C T 0 11 0 F T T A C C T C C T 0 9 18 G T T A T T T C T T 0 4 0 H T T A T T T C C T 0 0 12 I T T G T T T T C T 0 5 0

Appendix 8 229 Appendix 8

Dominance matrix based on agonistic interactions among males in three semi-free ranging eastern grey kangaroo (Macropus giganteus) populations: (a) Population A, (b) Population B, and (c) Population C.

(a) MALE Pu YB YR MW Gpi Des Losses Wins Purple 2 1 0 0 0 3 29 YB 8 2 0 0 0 10 20 YR 7 0 0 0 0 7 24 MW 1 3 4 0 0 8 8 Gpi 13 8 13 8 0 42 0 Des 0 7 4 0 0 11 0 TOTAL 29 20 24 8 0 0 76

Appendix 8 230

Blue Blue (b) MALE Green Orange Red Yellow Pink Purple White Snr Jnr Losses Wins Green 0 0 0 0 0 0 0 0 0 71 Blue Snr 6 0 5 2 0 0 0 0 13 42 Orange 5 6 0 1 0 0 0 0 12 13 Red 16 8 2 6 0 0 0 0 32 35 Yellow 8 0 1 2 0 2 0 0 13 43 Pink 9 5 0 4 3 4 0 0 25 5 Purple 10 7 4 2 10 0 0 2 35 16 Blue Jnr 4 5 5 5 5 1 1 0 26 0 White 13 11 1 17 16 4 9 0 71 2 TOTAL 71 42 13 35 43 5 16 0 2 205

Appendix 8 231

(c) MALE Green Blue Red Yellow Orange Purple Losses Wins Green 0 0 0 0 0 0 76 Blue 16 2 9 19 15 61 7 Red 2 0 1 20 0 23 11 Yellow 21 0 1 0 0 22 20 Orange 17 5 7 7 7 43 39 Purple 20 2 1 3 0 26 22 TOTAL 76 7 11 20 39 22 102 Appendix 9 232 Appendix 9

Allele frequencies for 10 microsatellite loci for three semi-free ranging eastern grey kangaroo (Macropus giganteus) populations.

Alleles at locus T19-1 Population 111 115 119 123 127 131 135 139 143 149 153 157 161 165 A - 0.144 0.087 0.077 0.029 0.048 - 0.019 0.077 0.269 0.077 0.038 0.019 0.115 B 0.042 0.083 - 0.125 0.063 0.188 - - - 0.188 0.271 - 0.021 0.021 C 0.119 0.119 0.024 0.095 0.214 - 0.167 0.024 0.024 - 0.143 0.071 - -

Alleles at locus G16-1 160 162 164 166 168 170 172 174 176 178 180 182 184 186 A 0.038 - - 0.104 0.189 0.104 0.179 0.255 - 0.047 0.028 0.019 0.028 0.009 B 0.042 0.125 0.063 - 0.042 0.063 0.208 0.375 - 0.063 - 0.021 - - C 0.143 0.048 - 0.071 - 0.095 0.071 0.190 0.071 0.286 - - 0.024 -

Alleles at locus G26-4 253 257 261 265 269 273 277 281 285 289 293 309 317 A 0.029 0.010 0.067 0.019 0.192 0.048 0.010 0.019 - - 0.067 - 0.067 B 0.125 - - 0.313 0.021 - - - - - 0.042 - - C - 0.114 0.114 0.023 0.341 0.023 - - 0.023 0.023 0.068 0.045 -

Appendix 9 233 Alleles at locus G26-4 continued 321 329 333 337 339 341 343 347 351 353 359 365 369 A 0.077 0.010 0.048 - - 0.019 0.038 - 0.029 0.038 0.106 0.010 0.096 B 0.021 - 0.292 - - - - 0.083 0.021 0.042 0.042 - - C - 0.045 - 0.023 0.023 - - - - 0.068 0.068 - -

Alleles at locus G31-1 113 115 117 119 121 123 125 127 129 131 133 A - - 0.356 0.298 - 0.058 0.096 0.067 0.087 0.019 0.019 B - - 0.438 0.125 0.021 0.167 0.021 0.104 0.021 0.104 - C 0.114 0.023 0.091 0.136 0.318 0.068 0.068 0.023 - 0.159 -

Alleles at locus T46-5 150 154 158 162 166 168 170 172 174 A 0.009 0.009 0.179 0.330 0.132 - 0.160 - 0.179 B - 0.021 0.042 0.313 0.396 - 0.042 - 0.188 C - - 0.136 0.250 0.432 0.023 0.136 0.023 -

Alleles at locus Pa595 196 200 204 210 212 216 220 224 228 232 236 240 244 248 252 A 0.031 0.010 0.052 0.083 0.083 0.031 0.167 0.031 0.073 0.094 0.083 0.115 0.094 0.010 0.042 B - - 0.021 0.021 0.104 - 0.208 0.021 0.146 0.063 0.313 0.083 - - 0.021 C - - - 0.050 0.125 0.050 - 0.025 0.200 0.250 0.225 - 0.025 0.050 -

Appendix 9 234

Alleles at locus T31-1 111 117 119 121 123 125 127 129 137 A 0.019 0.346 0.240 0.163 0.087 0.010 0.087 0.019 0.029 B 0.021 0.146 0.146 0.313 0.104 - - - 0.271 C 0.136 0.136 0.295 0.318 0.023 0.023 - 0.023 0.045

Alleles at locus G20-2 143 145 147 149 151 153 155 157 166 A 0.058 0.202 0.077 0.058 0.202 0.279 0.038 0.087 - B - 0.167 0.104 - 0.146 0.458 0.042 0.083 - C 0.023 0.023 - 0.318 0.409 0.114 0.068 - 0.045

Alleles at locus T32-1 154 160 162 164 166 168 170 172 174 176 178 180 182 184 A - 0.087 0.077 0.192 0.135 0.067 0.048 0.115 0.067 0.125 - 0.038 0.048 B 0.021 0.083 - 0.208 0.521 - - 0.104 0.063 - - - - - C 0.045 0.045 0.023 0.068 0.273 - 0.250 0.159 - 0.045 0.023 - - 0.068

Alleles at locus T3-1T 147 151 155 159 163 167 171 175 179 183 187 191 203 A 0.019 - 0.077 - 0.115 0.019 0.077 0.106 0.019 0.115 0.029 0.010 0.019 B 0.042 - - - 0.063 - - 0.104 0.021 - - 0.063 0.146 C - 0.091 0.068 0.182 0.114 - 0.023 - 0.114 0.068 - 0.091 -

Appendix 9 235

Alleles at locus T3-1T continued 211 215 221 227 239 243 255 259 261 263 265 285 289 A 0.019 0.048 - 0.010 0.058 0.019 0.010 0.010 - 0.067 0.058 0.029 0.067 B - 0.396 - - - - 0.063 0.021 - 0.042 - 0.021 0.021 C 0.068 0.023 0.023 - - - - - 0.068 - 0.045 - 0.023 Photo gallery 236

Photo Gallery

Photo gallery 237

Photo gallery 238

Photo gallery 239

Exciting discovery of the early arrival of pouch young in tammar wallabies (Macropus eugenii) on the Abrolhos Islands, Western Australia

Photo gallery 240

The rewards of fieldwork in remote locations - Cath Herbert aboard The Rat Patrol in the Abrolhos Islands, Western Australia.

Photo gallery 241

An exhausting journey, but well worth it. Mark Eldridge and Emily Miller after walking over a reef to check traps on West Wallabi Island, Western Australia.

Photo gallery 242

For the male tammar wallabies (Macropus eugenii) on the Abrolhos Islands, Western Australia: a bargain at any price!