The Application of Molecular Genetics to the

Conservation Management of Quolls, Dasyurus Species

(Dasyuridae:Marsupialia)

Karen B. Firestone

Ph. D.

Submitted December 1999 Abstract: The quolls are among the largest of the remaining carnivorous marsupials in the

Australasian region, and thus occupy an important ecological niche as top predators and scavengers. All quolls are currently in decline and threatened to some degree yet the application of molecular information to the conservation and management of quolls has been unexplored until now. In this thesis I use two independent and highly variable genetic marker systems, the mitochondrial DNA (mtDNA) control region and nuclear microsatellites, to explore various aspects of conservation genetics relevant to the management of quolls. These aspects include an examination of the phylogenetic or evolutionary relationships among all six species of quolls, an examination of the genetic diversity within populations and the degree of differentiation between populations of the four Australian species of quolls, and the definition of units for conservation within these species.

The development of suitable nuclear markers was a vital first step in defining levels of genetic variability and differentiation within and between the different populations and species. These markers proved to be highly variable and provided a wealth of information of relevance to the conservation of these species, and will be extremely useful in further studies.

The use of the mtDNA control region for phylogenetic analyses was a novel approach to examining this question in quolls and also proved to be highly informative. Results from these phylogenetic analyses highlight the necessity of 1) examining more than one exemplar of each species, as well as 2) finally bringing some consensus to the question of the evolutionary relationships among quolls. Results show that northern quolls form the earliest split from all other quolls and that western quolls are closely related to the two New Guinean species.

ii Furthermore, there is evidence for distinct lineages within species, corresponding to geographically separate or isolated populations.

Levels of genetic variability within populations were examined using the microsatellites developed previously. Genetic variation was significantly higher in western quolls than in any other species. This was surprising given the long term and widespread decline of this species.

There were also significant differences between populations within species in the level of genetic variability. Low levels of variability were usually found in small or captive bred populations or populations in severe decline.

Genetic differentiation between populations was also explored using microsatellites. Significant differentiation in allele frequency distributions was found between most pairwise population comparisons, indicating that each of these populations forms a separate management unit

(MU) for conservation purposes. One notable exception was found among populations of tiger quolls from a highly localized area in the Barrington Tops region of New South Wales.

Using microsatellites, these populations were not significantly subdivided and thus appeared to be one MU. Using mtDNA, however, these populations were significantly subdivided and thus should be considered separate MUs. Differences in the way these two genetic markers are inherited (mtDNA is maternally inherited, microsatellites are biparentally inherited) provides a clue as to the social structure and organization of these cryptic nocturnal species.

Consequently, the use of different genetic marker systems shows that there is sex-biased migration within this species.

iii Finally, the degree of genetic differentiation observed within tiger quolls does not conform to the currently recognized subspecific categories within this species. The major genetic split occurs between the Tasmanian and mainland populations of tiger quolls, not between

Dasyurus maculatus maculatus and D. m. gracilis. Thus, the Tasmanian and mainland populations form two distinct evolutionarily significant units (ESUs) for conservation purposes, and I propose that the Tasmanian populations should be elevated to the subspecific status to account for this.

iv Table of Contents

Abstract ...... 11

Table of Contents ...... v

Acknowledgments ...... viii

List of Figures ...... xiii

List of Tables ...... xv

Chapter 1. Introduction, Background, and Theoretical Framework ...... 1 Organization of This Thesis ...... 1 Description of Quolls ...... 1 History of European Discovery, Nomenclature, and Taxonomic Status ofQuolls ...... 5 Fossil Dasyures ...... 9 Current Classification of the Dasyures ...... 1 0 Conservation Status ...... 11 Population Declines, Range Contractions, and Current Distribution of Quolls ...... 13 Eastern Quoll, Dasyurus viverrinus ...... 14 Tiger Quoll, Dasyurus maculatus ...... 1 7 Western Quoll, Dasyurus geoffroii ...... 20 Northern Quoll, Dasyurus hallucatus ...... 22 Conservation Biology ...... 23 Molecular Genetics in Conservation ...... 24 Phylogeny of the Quolls ...... 25 Genetic Differentiation, Subspeciation, Conservation Units, and the 'Agony of Choice' ...... 27 Genetic Variability...... 32 0 bj ecti ves and Thesis Aims ...... 3 4 References ...... 3 5

Chapter 2. Isolation and Characterization of Microsatellites from Carnivorous Marsupials (Dasyuridae:Marsupialia) ...... 55 Introduction ...... 55 Materials and Methods ...... 55 Results and Discussion ...... 58 References ...... 60 Acknowledgments ...... 61

v Chapter 3. Phylogenetic Relationships Among Quolls Revisited: the mtDNA Control Region as a Useful Tool ...... 62 Introduction ...... 63 Review of Past Phylogenetic Reconstructions of Quolls ...... 64 The Control Region as a Phylogenetic Marker ...... 70 Materials and Methods ...... 71 Samples and DNA Extractions ...... 71 Control Region Amplification, Screening, and Sequencing...... 75 Statistical Analyses ...... 76 Results...... 77 Discussion ...... 88 Features of the Control Region in Dasyurus ...... 88 Phylogenetic Relationships Within Dasyurus ...... 89 Phylogenetic Relationships Within Species ...... 92 References...... 93 Acknowledgments ...... 10 1

Chapter 4. Variability and Differentiation: Microsatellites in the Genus Dasyurus and Conservation Implications for the Large Australian Carnivorous Marso pials ...... 103 Introduction ...... 104 Materials and Methods ...... 109 Study Populations, DNA Samples, and Population Screening ...... 109 Statistical Analyses ...... Ill Results ...... 114 Genetic Variability of Microsatellites in Quolls ...... 114 Deviations from Hardy Weinberg Equilibrium ...... 114 Genetic Diversity Among Species ...... 115 Genetic Diversity Among Populations Within Species ...... 118 Genetic Differentiation Among Species: Unique Alleles ...... 119 Genetic Differentiation Among Species: Allele Frequencies ...... 119 Genetic Differentiation Among Populations Within Species: Population Subdivision ...... 122 Genetic Differentiation Among Populations Within Species: Genetic Distance Measures ...... 122 Geographic Distance vs. Fst ...... 124 Assignment Tests ...... 126 Discussion ...... 128 Hardy-Weinberg Equilibrium...... 128 Genetic Diversity Among and Within Species ...... 128 Genetic Differentiation Among and Within Species ...... 130 Conservation Implications...... 131 Western Quolls ...... 131 Tiger Quolls ...... 132 Eastern Quolls ...... 133 Northern Quolls ...... 134 References ...... 136 Acknowledgments...... 146

vi Chapter 5. Phytogeographical Population Structure of Tiger Quolls Dasyurus maculatus (Dasyuridae:Marsupialia) an Endangered Carnivorous Marsupial ...... 148 Introduction ...... 149 Materials and Methods ...... 152 Samples and DNA Extractions ...... 152 Control Region Amplification, Screening, and Sequencing ...... 152 Microsatellite Amplification and Screening ...... 154 Statistical Analyses ...... 15 5 Results ...... 15 7 Control Region Variation ...... 15 7 Geographic Distribution of Control Region Haplotypes ...... 160 Phylogeographic Relationships of mtDNA Haplotypes ...... 160 Control Region Population Differentiation ...... 162 Micro satellite Variation ...... 166 Microsatellite Population Differentiation ...... 167 Discussion ...... 170 Within-Population Levels of Genetic Diversity ...... 170 Among-Population Levels of Genetic Differentiation ...... 171 Evolutionary Distinctions and Subspeciation Among Tiger Quolls ...... 173 Conservation Status, Management Issues, and Recommendations ...... 17 4 References ...... 17 6 Acknowledgments ...... 185

Chapter 6. Summary of Results and Conservation Implications for Quolls ...... 186 Introduction ...... 186 Summary of Results ...... 186 Conservation Implications of Relevance to Management of Quolls ...... 190 Areas for Further Study ...... 192 References...... 193

Appendices Appendix 1. Cloned microsatellite sequences ...... 195 Appendix 2. Voucher data for all samples used in these studies ...... 196 Appendix 3. Microsatellite genotypes of all quolls analysed in these studies ...... 205 Appendix 4. The mtDNA control region alignment...... 213 Appendix 5. Allele frequency distributions for microsatellite loci in four species (20 populations) of quolls ...... 221 Appendix 6. Allele frequencies for six microsatellite loci/nine populations of tiger quolls ...... 227 Appendix 7. Appendix 7. Hardy-Weinberg Equilibrium probabilities for all populations estimated by the Markov chain method for loci with more than five alleles ...... 228

vii Acknowledgments

As with any large project, many people helped me over the years; I owe a huge debt of gratitude to all of those who were instrumental in assisting me during the course of this work.

First and foremost, I would like to thank my supervisors Jack Giles, of the Zoological Parks

Board of New South Wales, Bill Sherwin from the University of New South Wales, and Chris

Dickman from Sydney University. Without their continued intellectual encouragement and tireless editing skills this thesis would not be in the form it is in. I also owe a great deal of thanks to Bill, who took a punt on a prospective student who had not even heard of the

Polymerase Chain Reaction when she started this project, yet never-the-less, trustingly provided the laboratory space and equipment to carry out this work. Jack Giles provided support in innumerable ways: he initially proposed the idea of this project and secured the first funding to begin this work. In addition, he allowed me the freedom to pursue a Ph D within the framework of my responsibilities as a Research Officer of the Zoological Parks Board and the logistic support of a wonderful working environment in which to do the majority of the analyses and writing. Where else would one get to look out from one’s office window at kookaburras and water dragons, or hear the loud thump of possums on the roof of your office during those late evenings, or listen to the raucous calls of ring-tailed lemurs, macaws, and snow leopards—all while overlooking beautiful Sydney harbour. Chris Dickman also provided support in many ways. When the going got rough in the field, Chris always bucked up my spirits by reminding me that his species were even rarer than mine! Chris’ constant encouragement whenever I had doubts helped me keep going on numerous occasions.

viii I also owe a great deal of thanks to all the wonderful people around Barrington who helped me through the last few years. Brad and Jenny Lewis, the owners of Barrington Guest House, and their family, fed me, housed me, generally looked after me, and let me have the run of the place. The staff and management at the Guest House kindly put up with me and always seemed highly bemused by all the weird things going on; these marvellous people often went out of their way to help me. I will always treasure the many memories from the Guest House.

Darcy and Maree Longbottom and their family have become a second family to me, Bill and

Kathy Dowling shared their enthusiasm and vast knowledge of the Australian bush (and also provided a ‘field base’ at their lovely home) and Dennis and Wendy Burt have become close friends. Moira Munroe of Nowendoc and many local community members from Copeland willingly helped me out. Without their long-term support, this project would not have been feasible.

This work also would not have been accomplished with so much fun without the great deal of help given to me by the many field assistants who accompanied me to Barrington. Mark

Caddey, Rocky De Nys, Sue Greig, Warwick Greville, Bronwyn Houlden, Lotem Levitan,

Matthew Milgate, Ross Knowles, Michael Pennay, Scott Priestly, Debbie Saunders, Kathy

Saunders, Eva Twarkowski, Sharon Warne, Jennifer White, and Phil Zammit all put up with rotten chicken and maggots for the opportunity to come help. I thank Eva and Scott for one particularly delirious, yet memorable week up at Barrington. A very special thank you to my friend Bob Longbottom who helped with many field trips and provided hours of amusement out in the bush.

ix I also thank my lab mates, both past and present, without whose assistance I would probably still be doing PCRs even now. Warwick Greville, Andrea Taylor, Bronwyn Houlden, Phillip

England, Deryn Alpers, Deirdre Sharkey, Michael Kreutzen, Elena Valsecchi, Angie Penn, and Ayesha Penny, all were instrumental in providing answers to my questions and much needed assistance when things were not working in the lab, which was quite often! In particular, I thank Deirdre for all the reassuring conversations over the past few months—her constant encouragement has meant a great deal to me. Phillip England provided instruction on manufacturing and screening the first genomic library presented in Chapter 2. Dave Edwards trialed the resulting microsatellite primers on Antechinus species. To Martin Elphinstone,

Dean Jerry, and Margaret Heslewood, I also owe huge thanks. Martin and Dean provided the instruction and assistance with screening large numbers of samples using their TGGE system over a condensed two-week flying visit to their lab at Lismore. Margaret provided great help with phylogenetic analyses and the use of PAUP. In addition, Luca Fumagalli was kind enough to provide the mtDNA primers and numbat and bandicoot sequences (before they were published) and Joe Zuccarello provided assistance with the Kishino-Hasegawa test used in Chapter 3. Eli Geffen was instrumental in providing a great deal of assistance with statistical analysis of microsatellites used in Chapter 4, in particular he provided the analysis for

Figures 15, 17, 19, part of Figure 16, and Table 7. Bronwyn Houlden always provided an answer when I had a question regarding either laboratory procedures or analyses—her help was immeasurable.

Many colleagues and institutions provided samples used in these studies. I thank Chris

Belcher, Scott Burnett, Joyce Byron, Nora Cooper, Frank Craven, Ronnie Darr, Russ

x Dickens, Joan Dixon, Bill Dowling, Lina Frigo, Cliff Gallagher, Linda Gibson, Jack Giles, Jim

Griffiths, Graham Hall, Menna Jones, Androo Kelly, Brad Lewis, Darcy Longbottom, Luke

Leung, Keith Morris, Dianne Moyle, M. Murray, Meri Oakwood, Michael Pennay, Lisa

Pope, Scott Priestly, Judy Rainbird, Bruce Read, John Seebeck, Jeff Titmarsh, Larry

Vogelnest, Brad Walker, The Australian Museum, The Western Australian Museum, The

Museum of Victoria, the Queen Victoria Museum, Featherdale Wildlife Park, Trowunna

Wildlife Park, Perth Zoo, and Taronga Zoo. I also thank State Forests of New South Wales and the NSW National Parks and Wildlife Service who provided permits to work on lands under their tenure.

Other colleagues provided critical readings of various chapters before they went to publication. Along with Jack Giles, Chris Dickman, and Bill Sherwin who read each chapter,

Carey Krajewski, Mike Westerman, Patrick Luckett, Mark Springer, and Chris Quinn provided comments on Chapter 3; Eli Geffen, Mike Roy, and Bronwyn Houlden, provided comments on Chapter 4; Bronwyn Houlden, Josh Ginsberg, Louis Bernatchez, and two anonymous reviewers provided critiques on Chapter 5.

I also would like to thank my colleagues at Taronga Zoo who have helped me in numerous ways. Elizabeth Cook, Cliff Gallagher, Jodie Ible, Ed Lonnon, and Greg Webb kindly provided assistance when called upon.

Funding for this work was received from an Australian Postgraduate Research Award, the

Estate of W.V. Scott, Barrington Guest House, the Zoological Parks Board of New South

Wales, State Forests of New South Wales, and Cassegrain Wines.

xi Finally, I owe a very special debt of gratitude to Warwick Greville who helped me out in so many ways, I do not know where to start. He helped me in the field, the lab, and at home. I can not express how much your love and support has meant to me over the years and I look forward to our continued collaboration on our next joint project: the birth of our child.

xii List of Figures

Figure 1. The six different species of quolls ...... 4

Figure 2. Hind foot in two species of quolls...... 5

Figure 3. Classification of the Dasyures...... 11

Figure 4. Current (black) and former (grey + black) distribution of quolls in Australia ...... 15

Figure 5. Partial phenogram of dental similarity ...... 65

Figure 6. Cladistic relationships of the Australian quolls based on allozyme electrophoresis...... 67

Figure 7. Cladistic relationships of the quolls based on dental, basicranial, and morphological characters...... 67

Figure 8. Majority rule consensus tree based on complete cytochrome b, 12S rRNA, and protamine P1 sequences...... 69

Figure 9. Phylogenetic relationships among Dasyurus based on molar morphology ...... 70

Figure 10. Parsimony informative sites of 30 control region sequences of quolls and the Tasmanian devil...... 78

Figure 11. Majority rule consensus tree of 33 equally most parsimonious trees, based on 148 parsimony informative sites in the mtDNA control region ...... 80

Figure 12. Neighbor-joining tree based on pairwise Tamura distances ...... 82

Figure 13. Minimum spanning network based on the number of base pair substitutions between pairs of sequences...... 87

Figure 14. Current (black) and past (grey + black) distribution of the Australian quolls including sample sites...... 106

Figure 15. Estimated cumulative number of alleles per sample size for each species by Monte Carlo simulation...... 116

xiii Figure 16. Mean and standard error of corrected allelic diversity (A’) and

expected heterozygosity (HE) at six microsatellite loci (A) among species and (B) within species of quolls...... 117

Figure 17. Estimated number of unique alleles for each species by Monte Carlo simulations...... 120

Figure 18. MDS based upon Euclidean distances among 15 populations of quolls .... 121

Figure 19. Neighbor-joining tree of 15 populations of quolls using Nei’s D ...... 125

Figure 20. Map of Australia showing the current (black) and past (black plus gray) distribution of D. maculatus, the distribution of the subspecies, and sampling localities...... 151

Figure 21. Sequence variation among 12 mtDNA control region haplotypes detected among 93 individual tiger quolls ...... 158

Figure 22. Phylogenetic relationships among the mtDNA haplotypes ...... 161

Figure 23. Minimum spanning network of 12 different control region haplotypes found in tiger quolls...... 163

xiv List of Tables

Table 1. Conservation status of the Australian quolls ...... 12

Table 2. Six primer pairs used to characterize genetic variability and differentiation within Dasyurus species...... 59

Table 3. Samples used in this study...... 73

Table 4. Pairwise Tamura distance (below diagonal) and total number of observed substitutions (pairwise deletion method; above diagonal) ...... 84

Table 5. Kishino-Hasegawa (1989) test results...... 86

Table 6. Species and populations sampled...... 110

Table 7. The number and percentage (in parenthesis) of unique alleles between four species of quolls...... 121

Table 8. Pairwise comparison of Nei’s D (below the diagonal) and FST (above the diagonal) ...... 123

Table 9. Species assignment test...... 126

Table 10. Population assignment test for 15 populations of quolls ...... 127

Table 11. Subspecies, sample populations, locations, and sample sizes used in this study) ...... 153

Table 12. Haplotype frequencies, haplotype diversity, and nucleotide diversity in nine tiger quoll populations...... 159

Table 13. Nested analysis of molecular variance for mtDNA in tiger quolls ...... 165

Table 14. Microsatellite diversity in tiger quolls averaged over six loci ...... 167

Table 15. Pairwise measures of microsatellite differentiation in nine populations of tiger quolls...... 168

Table 16. Pairwise population subdivision of nuclear loci ...... 169

xv Chapter 1. Introduction, Background, and Theoretical Framework

Organization of this Thesis

This thesis is presented as a series of four papers together with an introductory chapter and a summary chapter. The first chapter presents an introduction to the species under study, the members of the genus Dasyurus, or quolls, as well as the theoretical framework uniting the following papers. The following chapters are all papers that have been published or submitted for publication in international journals. The second chapter is a description of the isolation and characterization of the nuclear microsatellites which were utilized in the following papers

(Firestone 1999). The third chapter presents an analysis of the evolutionary relationships among all species of quolls using the mitochondrial DNA (mtDNA) control region as a novel locus (Firestone in press). The fourth chapter is a detailed examination of the genetic variability and differentiation found within and among the four Australian species of quolls using the microsatellite markers developed in the first paper (Firestone et al. submitted). The fifth chapter presents an examination of the conservation units found within tiger quolls, Dasyurus maculatus, and provides genetic evidence for a new subspecies within this species (Firestone et al. 1999). The concluding chapter presents an analysis and synthesis of the findings in the previous papers, recommended conservation actions for quolls, and suggestions for future areas of research.

Description of Quolls

Quolls are members of the family Dasyuridae, a large family of approximately sixty species ranging in size from around 4g (Planigale ingrami) up to 8kg (male Sarcophilus

1 harrisii). The dasyurids are found in Australia and New Guinea with additional populations restricted to a few other islands. Quolls are medium- to large-sized carnivorous or insectivorous marsupials. Early descriptions indicated the broad resemblance of quolls to some members of the Carnivora (weasels and polecats), in their elongated visage and streamlined body (e.g. Phillip 1789). Other early descriptions took more note of their mode of reproduction and placed them with the opossums, (e.g. Shaw 1800). Quolls all possess four or five separate toes on the hind foot (didactyly), in contrast to the peramelids and diprotodonts which have a fused second and third toe (syndactyly). In addition, quolls are polyprotodonts, having four pairs of upper and three pairs of lower incisors. Quolls also possess four pairs of upper and lower molars, two pairs of premolars, and very well developed canines. Their dentition is developed to a greater or lesser degree for carnivory or insectivory.

Externally, quolls are easily distinguished from any other species in Australia or New Guinea by their spotted pelage (Figure 1). The four Australian quolls are readily distinguished from each other by the pattern of spots, the presence or absence and size of the hallux (the clawless first toe on the hind foot, Figure 2), by geographical location (Figure 4), and by size and coat color. All quolls are sexually dimorphic with males being larger than females. Tiger quolls

(spotted-tailed quolls), Dasyurus maculatus, are the largest of the genus with males weighing up to 7 kg and females up to 4 kg (Edgar and Belcher 1995), and are the only quolls with spots extending onto the tail; they possess a hallux (Figures 1A, 2A). Eastern quolls,

Dasyurus viverrinus, are distinguishable from all other quolls by the lack of a hallux; they are a medium sized

2 quoll (average weights: males 1300 g, females 880 g), they do not have spots on the tail, and the tail tends to taper to a whitish tip at the end (Figures 1B, 2B; Godsell 1995).

Northern quolls, Dasyurus hallucatus, are the smallest of the Australian group (average male and female weights 760 g and 460 g, respectively; Oakwood 1997), they do not have spots on the tail, but do possess a hallux (Figure 1C). Western quolls or chuditch, Dasyurus geoffroii, are also a medium sized quoll (average weights: males 1310 g, females 890 g), but can be distinguished from eastern quolls by the presence of a hallux, and a tail that tapers to a black tip at the end (Figure 1D; Serena and Soderquist 1995).

The two New Guinean species are readily differentiated from each other by size, presence or absence of striate pads on the feet, the size of the hallux, and geographical location in Papua

New Guinea. The bronze quoll, Dasyurus spartacus, is comparable in size to eastern or western quolls, lacks striations, has a relatively smaller hallux, and lives in low mixed savannah in southwest Papua New Guinea (Figure 1E; Van Dyck 1987; Flannery 1995). The New

Guinea quoll, Dasyurus albopunctatus, is a small animal, comparable in size to the northern quoll, has striated foot pads, a relatively larger hallux than D. spartacus, and lives in rainforest

(Figure 1F).

3 Figure 1. The six different species of quolls.

A. Tiger (Spotted-tailed) quoll, D. maculatus. B. Eastern quoll, D. viverrinus. Note Largest of the quolls, spots extend onto tail, whitish tip and lack of spots on tail. Has has a hallux. no hallux.

C. Northern quoll, D. hallucatus. Smallest D. Western quoll (chuditch), D. geoffroii. of the Australian quolls, has a hallux, no Differentiated from D. viverrinus in having a spots on tail. hallux and black tail tip.

E. Bronze quoll, D. spartacus. From PNG. F. New Guinea quoll, D. albopunctatus. From Medium sized animal, with a relatively small PNG. Small animal, relatively large hallux, hallux, no striations on foot pads. striations present on foot pads.

A-D Used with permission of Nature Focus; E-F Used with permission of P. Woolley, D. Walsh.

4 Figure 2. Hind foot in two species of quolls.

A. Hind foot of D. maculatus showing the B. Hind foot of D. viverrinus without hallux the hallux

History of European Discovery, Nomenclature, and Taxonomic Status of Quolls

Since their first discovery by Europeans in the early 1770s, the scientific nomenclature and taxonomic status of quolls has undergone many revisions (Waterhouse 1846; Pocock 1926;

Iredale and Troughton 1934; Tate 1947; Kirsch and Calaby 1977; Mahoney and Ride

1988). The earliest record of European discovery of quolls was from 1770. What is now known as the northern quoll (see Mahoney and Ride 1984), was described by the local aborigines as ‘jaquol,’ ‘taquol’ (Parkinson 1773) ‘je-quoll’ (Banks 1896), or ‘quoll’

(Hawkesworth 1773) to the early explorers.

5 In 1789, a more detailed description of a tiger quoll, D. maculatus, was published in Captain

Phillip’s journal; it was thought to be related to the eutherian mustelids and was called the

‘spotted martin’ (Phillip 1789). In 1792, this species was given the scientific name of

Viverra maculata (Kerr 1792) based on Phillip’s description. It was later placed in the genus Dasyurus as D. macrourus (Geoffroy 1803). However, Gray (1843) called the species Dasyurus maculatus. Matschie (1916), placed the species in the genus Dasyurops, while Pocock (1926) placed it in a new genus, Stictophonus. The taxonomic decision of

Tate (1947) holds. There are currently two recognized subspecies of D. maculatus (Tate

1952); the nominate form, D. m. maculatus, from the south east of the mainland and

Tasmania, and D. m. gracilis, initially described as D. gracilis from north Queensland

(Ramsay 1888; Tate 1952).

The earliest known description of the eastern quoll, D. viverrinus, was as the ‘spotted opossum’ (Phillip 1789). White (1790) considered this species to be a variety of the brush- tailed phascogale, Phascogale tapoatafa, “…only differing from the Tapoa Tafa in its external colour, and in being spotted.” The eastern quoll was later described more fully and given the scientific name Didelphis maculata (Kerr 1792). In 1799, this species was placed within the genus Dasyurus as D. maculatus (Lacépède 1799), but Shaw considered it to be part of the genus Didelphis and erected the species name viverrina (Shaw 1800).

Geoffroy named the black form D. viverrinus and the grey form D. maugei (Geoffroy

1804). Gould, however, determined that D. maugei and D. viverrinus “were the same species, a fact which he ascertained by finding in the same

6 litter both the black and grey varieties” (Gould 1841). There are currently no recognised subspecies of D. viverrinus.

The nominate form of the western quoll or chuditch, D. geoffroii geoffroii, was described originally as D. geoffroii (Gould 1841). This specimen was from Liverpool Plains in New

South Wales. The western subspecies, D. g. fortis, was distinguished from D. g. geoffroii on external morphological characters such as its larger size, subtle differences in coat and skin color, and on differences in skull morphology such as larger bullæ (Thomas 1906). These two subspecies were placed in the genus Dasyurinus by Matschie (1916), and in a new genus,

Notoctonus, by Pocock (1926), then later repositioned within the genus Dasyurus (Ride

1970).

Northern quolls, D. hallucatus, were the first of the quolls to be given a scientific name:

Mustela quoll (Zimmermann 1777; Mahoney and Ride 1984), again indicating the general similarity of this species to the mustelids. Later this species was described as D. hallucatus

(Gould 1842). Thomas characterised three other races: D. h. predator, D. h. exilis, and D. h. nesæus which were distinguished from D. h. hallucatus and each other on geographic locality, external morphological characters such as coat colour and size, and skull characters such as width of nasal bones (Thomas 1909; Thomas 1926). Pocock (1926) put these taxa in the genus Satanellus; Ride (1970) replaced them in the genus Dasyurus. There has been much debate over the past twenty years on the generic status of this species with some authors suggesting northern quolls belong in Satanellus (Pocock 1926; Archer 1982; Kirsch

7 and Archer 1982; Marshall et al. 1990) , while others include it in Dasyurus (Baverstock et al. 1982; Van Dyck 1987;

Baverstock et al. 1990; Kirsch et al. 1990; Krajewski et al. 1994; Wroe and Mackness

1998). Little attention currently is paid to the subspecific status of this species, however, and these trinomials are not referred to in the literature or in management decisions today

(Maxwell et al. 1996; Strahan 1998).

The New Guinea quoll, D. albopunctatus, was described by Schlegel (1880). Three subspecies were considered to be within this species: the nominate form, D. a. albopunctatus (Schlegel 1880), D. a. fuscus (Milne-Edwards 1880), and D. a. dæmonellus (Thomas 1904). These three subspecies were differentiated from each other on size. Subsequently, however, these differences were thought to be due to sexual dimorphism in the case of albopunctatus and dæmonellus and due to age differences for albopunctatus and fuscus and all subspecies were therefore synonymised to albopunctatus (Tate 1947).

It was not until recently that the latest species, D. spartacus, the bronze quoll, was described

(Van Dyck 1987). D. spartacus was referred to previously as D. geoffroii because of its overall similarity to that species. On closer examination, however, this species was shown to be differentiated from all other dasyures on a number of classical morphological characters such as dental and basicranial structures, size, coat colour and size of spots, possession and size of the hallux, and size of the pinna (Van Dyck 1987). There are no recognised subspecies of D. spartacus.

8 Fossil Dasyures

Although the fossil record is poor, a few fossil forms of dasyures have been discovered and described; these are reviewed more fully in Archer (1982) and Wroe (1998). As noted in

Ride (1964), Dasyurus affinis was described by McCoy in 1865 from Quaternary fossil remains in Victoria. D. affinis was later synonymised with D. maculatus (Mahoney 1964).

Owen included an illustration of Dasyurus mordax from another Quaternary fossil fragment

(left dentary from P2-M4) found in an unidentified cave from an unknown location in Australia

(Owen 1877). However, the holotype, and only specimen, is missing and according to

Archer this specimen should be treated with suspicion since “Owen’s almost useless illustration … suggests a beast unlike any other species of Dasyurus” and is either diseased or possibly a fake (Archer 1982). In 1910, a new species was described from subfossils found on King Island in the Bass Strait as Dasyurus bowlingi (Spencer and Kershaw 1910).

The taxonomic distinction of this species was based mainly on differences in cranial length and shape of the auditory bullæ. Since the auditory bullæ are highly variable even within individuals, this distinction was deemed erroneous and D. bowlingi was recognised subsequently as conspecific with D. maculatus (Marshall and Hope 1973). Dasyurus dunmalli is described from Pliocene fossil remains from Queensland and New South Wales and is the only currently accepted fossil taxon (Bartholomai 1971).

Archer (1982) reviewed the known fossil dasyurids from the Miocene, Pliocene, Plio-

Pleistocene, and to some extent, the late Quaternary (Holocene). From the Pliocene D. dunmalli is represented in the Chinchilla Local Fauna of Queensland and the Bow Local

Fauna of New South Wales. Quolls are also represented in the Plio-Pleistocene

9 by one unaffiliated specimen from the Floraville local fauna of Leichhardt River in northwestern Queensland and by a specimen of Dasyurus hallucatus from the Fisherman’s

Cliff local fauna of southern New South Wales (Marshall 1973). There is also a larger

Dasyurus represented by several teeth in the Fisherman’s Cliff local fauna that could not be identified any further than genus; this animal is similar in size to D. geoffroii, D. viverrinus, or

D. dunmalli.

The occurrence of D. hallucatus in southern New South Wales Pleistocene deposits (Archer

1982) is well outside of its present known distribution. D. viverrinus is known from

Pleistocene deposits in southeastern Queensland (Archer 1978). This occurrence, too, is outside the known historical distribution of eastern quolls.

As yet, there are no Dasyurus specimens from the Oligocene/Miocene Riversleigh fauna.

The poor showing of fossil dasyurids is possibly due to smaller species being overlooked, or the relative scarcity of the larger species. Another possibility for the lack of many fossil forms is that the dasyures have undergone a relatively late and rapid radiation (Baverstock 1990;

Wroe 1996; Wroe, 1997; Wroe, 1999).

Current Classification of the Dasyures

The current higher order classification of the dasyures is based on Kirsch (1976), Archer

(1982), and Aplin and Archer (1987) and is shown in Figure 3. The nomenclature of the genus, species, and subspecies is after (Strahan 1998) for the extant Australian taxa and after

(Flannery 1995) for the extant New Guinean taxa.

10 There are currently six extant and one fossil species within the genus Dasyurus. Three of these species have associated subspecific taxa.

Figure 3. Classification of the Dasyures.

Supercohort: Marsupialia (Illiger 1811) Cohort: Australidelphia (Szalay 1982) Order: Dasyuromorphia (Gill 1872) Superfamily: Dasyuroidea (Goldfuss, 1820) Family: Dasyuridae (Goldfuss, 1820) Subfamily: Dasyurinae (Goldfuss, 1820) Tribe: Dasyurini (Goldfuss, 1820) Genus: Dasyurus (Geoffroy 1797) Species: maculatus (Kerr 1792) subspecies: maculatus (Kerr 1792) subspecies: gracilis (Ramsay 1888) Species: viverrinus (Shaw 1800) Species: geoffroii (Gould 1841) subspecies: geoffroii (Gould 1841)1 subspecies: fortis (Thomas 1906) Species: hallucatus (Gould 1842) subspecies: hallucatus (Gould 1842) subspecies: exilis (Thomas 1909)2 subspecies: nesæus (Thomas 1926)2 subspecies: predator (Thomas 1926)2 Species: albopunctatus (Schlegel 1880) Species: spartacus (Van Dyck 1987) Species: dunmalli (Bartholomai 1971)1 1 Taxon presumed extinct, 2 Uncertain taxonomic status

Conservation Status

Of the six species of quolls endemic to Australia and Papua New Guinea, there is virtually nothing known of the conservation status of the two New Guinean species (Dasyurus albopunctatus and D. spartacus), except that they are thought to be vulnerable (Flannery

1995). However, the Australian species of quolls have been severely affected in terms of range contraction and population decline and are given varying degrees of protection (Table

1).

11 Table 1. Conservation status of the Australian quolls.

Species/ Dasyurus maculatus Dasyurus viverrinus Dasyurus geoffroii Dasyurus hallucatus Authority 11 NSW1 Schedule 2: vulnerable Schedule 1, part 1: Schedule 1, part 4: na endangered presumed extinct ACT2 not listed Schedule 6: vulnerable Schedule 6: vulnerable na to or threatened with to or threatened with extinction extinction

TAS3 protected but not listed protected but not listed na na VIC4 endangered Schedule 2: presumed extinct, not listed na extinct SA5 Schedule 7: endangered Schedule 7: endangered Schedule 7: na species, presumed species, presumed endangered species, extinct extinct presumed extinct QLD6 D. m. maculatus , na Schedule 1: presumed Schedule 5: common Schedule 3: vulnerable extinct D. m. gracilis, Schedule 2: endangered12

NT7 na na Schedule 7: specially Schedule 1: protected, protected, presumed low risk extinct WA8 na na Schedule 1: rare or not listed likely to become extinct FED9 D. m. maculatus , Schedule 1, part 2: Schedule 1, part 1: not listed Schedule 1, part 2: vulnerable endangered vulnerable D. m. gracilis, Schedule 1, part 1: endangered12 IUCN10 D. m. maculatus , Low risk, near Vulnerable Low risk, near vulnerable threatened threatened D. m. gracilis, endangered12

1NSW=New South Wales National Parks and Wildlife Service; Threatened Species Conservation Act 1995. 2ACT= Department of Urban Services, Environment ACT; Nature Conservation Act 1980. 3TAS=Tasmanian Department of Environment and Land Management; the two species are protected under the general National Parks and Wildlife Act 1970, but not protected under the Threatened Species Protection Act 1995. 4VIC=Victorian Department of Natural Resources and Environment; Flora and Fauna Guarantee Act 1988, Wildlife Act 1975. 5SA=South Australian Department of Environment, Heritage, and Aboriginal Affairs; National Parks and Wildlife Act 1972. 6QLD=Queensland Department of Environment; Nature Conservation Act 1992, Nature Conservation (Wildlife) Regulation 1994. 7NT= Parks and Wildlife Commission of the Northern Territory, Territory Parks and Wildlife Conservation Act 1993. 8WA= Western Australia Department of Conservation and Land Management, Wildlife Conservation Act 1950. 9FED= Department of Environment, Environment Australia, Biodiversity Group; Endangered Species Protection Act 1992. 10IUCN=World Conservation Union, Red List. 11na=not applicable; the species never existed in that State or Territory. 12Queensland, the Federal Government, and the IUCN give the two subspecies of D. maculatus separate listings. Current at January 1998.

12 The various governmental departments, agencies and branches responsible for the listing of species for conservation purposes quite often have different levels of protection for these species or different terms for the same levels of protection. This variety and degree of protection often complicates the conservation status of the species in question. In addition, lists are continually being reviewed and the status of a species or subspecies is subject to change. Although all Australian species of quolls have declined and are protected on a state- wide basis, only three of the four species (D. maculatus, D. geoffroii, and D. viverrinus) have been given Federal protection (Table 1). Currently there is no special Federal protection for D. hallucatus, although there is strong evidence for broad scale declines in this species (Braithwaite and Griffiths 1994; Oakwood 1997).

Population Declines, Range Contractions, and Current Distribution of Quolls

The decline of the mainland dasyures is a complex issue and cannot be attributed to any one cause. Rather, it appears that a number of factors, acting in conjunction, may be implicated in the decline of each species. Factors such as the destruction of habitat (including the clearing of habitat for agriculture and urban development, and the reduction of vegetative cover by introduced herbivores such as sheep, cattle, and rabbits), competition with introduced predators (foxes, cats, and probably to a lesser degree, dogs), human persecution by hunting, poisoning or trapping, severe and prolonged drought, altered fire regimes, ingestion of poisonous prey (e. g. introduced cane toads, Bufo marinus), and the possibility of a disease epidemic may have affected one or more of these species (see reviews in Mansergh 1984;

Dickman and Read 1992).

13 Eastern quoll, Dasyurus viverrinus

Eastern quolls once occurred throughout much of the east coast and the south east of the mainland, some off-shore islands, and in Tasmania (Figure 4A). This species is now restricted solely to Tasmania where it is generally protected but is not given any special recognition as threatened or vulnerable wildlife (Table 1). This species was once very numerous and its decline has been well documented; accounts from the late 1800s tell of capture rates of eastern quolls that were a hundred times greater than that of the larger tiger quolls (see Mansergh 1984). Waterhouse noted that both the black and grey variety of D. viverrinus were common in New South Wales and Van Diemen's Land (Tasmania)

(Waterhouse 1846). In the Sunbury district of Victoria, Batey (1907) states that eastern quolls were “very numerous” around 1846, but later “they seemed to be infested with a burrowing maggot which brought them almost to the verge of extinction.” He also notes that they regained their numbers to some extent after that time and that up until 1875 they were

‘fairly numerous’ but from about 1887 they had not been seen in that area (Batey 1907).

The rapid decline of this species in South Australia was also remarked by Wood-Jones

(1923). He noted that this species’ range in South Australia was formerly very wide and that its “rapid decrease started about the year 1900, and during that and the two following years the so-called ‘common’ Native Cat practically disappeared from South Australia.” Victoria and New South Wales (except for the area immediately around Sydney) were affected similarly. Wood-Jones (1923) and others (see Morrison 1945) attributed the rapid decline of the species to an epidemic disease, although there is still no strong proof of this.

14 Figure 4. Current (black) and former (grey + black) distribution of quolls in Australia. A) Dasyurus viverrinus, B) Dasyurus maculatus, C) Dasyurus geoffroii, D) Dasyurus hallucatus. A, B, and C redrawn from Strahan (1995). D redrawn from Braithwaite and Griffiths (1994).

A B

C D

15 Eastern quolls were not safe on island refugia either, as they were “always more or less of a rarity” on Kangaroo Island (Wood-Jones 1923). In addition, by 1929 eastern quolls were reported as only being “seen occasionally” on Flinders and Cape Barren Islands in the Bass

Strait (Le Souef 1929). There is some contention, however, as to whether eastern quolls ever occurred on these islands and that the species seen was, in fact, the tiger quoll (Hope

1972). Despite the rapid decline observed, there is some indication that the species may have been sufficiently isolated from, or possibly somewhat resistant to, a disease epidemic in certain areas of Victoria (Fleay 1932; Fleay 1935). Nevertheless, there have been no specimen records of eastern quolls anywhere in Victoria since 1950 and no sightings of this species in the State since 1960 (Hampton et al. 1982). The relative isolation of certain areas in New South Wales (e. g. the spur of land leading to South Head in Sydney, and the

Illawarra region along the south coast) is thought also to be partly responsible for the survival of this species until the late 1960s or early 1970s.

Reasons given for the decline of eastern quolls are variable. A disease epidemic was suggested (Wood-Jones 1923) and it was further suggested that this may have been

Toxoplasmosis gondii (Shepherd and Mahood 1978) since the symptoms of this disease are similar to those that were reported in dying animals. However, no evidence of this protozoan was found in any samples of eastern quolls that were tested in Tasmania (Green 1967).

Other factors involved in the decline of this species may be related to habitat destruction, competition with and predation by introduced carnivores, and prolonged persecution by man

(Fleay 1932). Alternatively, the widespread poisoning

16 of rabbits earlier this century may have had a broad impact on eastern quoll numbers (J.

Giles, pers. comm.).

Although there are continued reported sightings of eastern quolls from other areas on the mainland of Australia, the last confirmed specimen was a female found as a road kill in

Nielsen Park, Vaucluse in the eastern suburbs of Sydney in 1963 (Australian Museum registration number M8382). This colony survived as an isolated stronghold until the lantana scrub was removed from the park in an attempt at bush regeneration. The removal of this dense cover apparently left no suitable habitat for the species. The lack of any definitive sightings or records from the mainland since 1963 indicates that eastern quolls currently must be ‘presumed extinct’ on the mainland. (See however, Robinson (1987) who states, without reference, that the last record of an eastern quoll on the mainland was in 1970 at Macquarie

Pass National Park in the Illawarra region of NSW and Caughley (1980) who says eastern quolls were present at Mt. Baldhead, Victoria in 1977.) Despite the apparent demise of this species on the mainland, it still survives in Tasmania where it is thought to be relatively secure.

It now occupies approximately 15% of its pre-European range (Braithwaite and Griffiths

1994; Figure 4A).

Tiger quoll, Dasyurus maculatus

Tiger quolls were once widespread over the south east of the mainland on both sides of the

Great Dividing Range. Their range spread into parts of Victoria, South Australia, and New

South Wales (Figure 4B). In addition a geographically separate population of the morphologically distinct subspecies, D. m. gracilis, is found in northern

17 Queensland. Tiger quolls are also present in Tasmania where they are given the same level of protection as the eastern quoll (i. e. generally protected, but not specifically listed; Table 1) even though they are not as numerous as eastern quolls (Jones 1995). Tiger quolls also once occurred in the Bass Strait (King, Flinders, Cape Barren, and Deal Islands) but were last seen on King Island in 1923 (see Hope 1972 for a review).

Although they were widespread, tiger quolls were never as numerous as their smaller relative,

D. viverrinus. In describing the fauna from the outskirts of Melbourne from the mid-1800s,

Batey noted that D. maculatus “had always been a rare animal…” particularly in relation to

D. viverrinus (Batey 1907). Wood-Jones noted that the tiger quoll “seems always to have been most numerous in the more wooded districts of the coast” but was probably never abundant in South Australia (Wood-Jones 1923). By 1932 Fleay also noted that both dasyures were rare in Victoria and that the tiger quoll was even rarer than the eastern quoll

(Fleay 1932). Sixteen years later, Fleay described the existence of tiger quolls as

“precarious” and stated further that “only in Tasmania may this spotted and primitive-looking creature be said to exist in any appreciable numbers” (Fleay 1948). Tiger quolls fared little better in NSW and were considered to be uncommon, although by this time they were more plentiful than eastern quolls (Marlow 1958).

Since European settlement, tiger quolls have had an overall range reduction of approximately

20% (Braithwaite and Griffiths 1994). However, some areas have been affected more severely than others (e. g. Victoria has had about a 50% reduction; Mansergh 1984) (Figure

4B) and this species is thinly dispersed over most of its

18 remaining range. The decline of this species probably occurred around the same time as that of eastern quolls and is attributed partly to the same factors as those affecting eastern quolls, i. e. habitat destruction, human persecution, introduced predators and an epidemic disease.

Inadvertent reduction by poisoning for dingoes, Canis familiaris, and rabbits, Oryctolagus cuniculus, may also have been a factor. In north Queensland, D. m. gracilis may have been affected by the introduction of cane toads, Bufo marinus, however it is unclear whether this is a primary cause for their decline (Burnett 1993; Burnett 1997).

Table 1 lists the conservation status of tiger quolls in each State throughout the species range.

D. maculatus is now thought to be extinct in South Australia; the last record is from 1958

(Archer 1979). However, it still occurs in other areas of the mainland, and in Tasmania. In

Victoria the species is restricted to a few fragmented populations, the strongholds being in southwestern Victoria, the Otway Ranges, and the upper Snowy River valley (see reference in Hampton et al. 1982; Mansergh 1984). In Tasmania the distribution of tiger quolls is considered patchy (Mansergh 1984) although it is not given formal listing. In Queensland D. m. gracilis is now classed as endangered under Commonwealth legislation and by the IUCN

(Maxwell et al. 1996). D. m. maculatus is classed as vulnerable in New South Wales and southern Queensland. The status of this species has been under review recently (Watt 1993), and is liable to change in a number of States.

19 Western quoll, Dasyurus geoffroii

Western quolls (chuditch) have been the most severely affected of all quolls since European settlement. Western quolls once ranged over 70% of the continent (Orell and Morris 1994); at that time, they were the most widespread of any of the quolls, being found in every mainland state. Burbidge et al. extended the known (former) distribution of the western quoll; their results indicated that this species may have been sympatric with D. hallucatus

(Burbidge et al. 1988). However, they are now restricted to approximately 5% of their former range (Braithwaite and Griffiths 1994; Orell and Morris 1994) in the extreme southwest of Western Australia (Figure 4C).

The type specimen, D. g. geoffroii, was recorded from Liverpool Plains in New South

Wales (Gould 1841); this is one of only three records of the species from this State (Dickman and Read 1992) and it has not been seen in the State since 1867 (Marlow 1958; Caughley

1980). Records for only three specimens from the Northern Territory exist (Johnson and

Roff 1982) and Parker noted that it was then either very rare or extinct in that State (Parker

1973). Wood-Jones was unsure how common the western quoll was in the South Australia, but noted only two records (Wood-Jones 1923); he considered that it then must have been regarded as extinct in South Australia unless it was still to be found in the northern part of the

State since “men who have been professionally interested in the fauna of the State for a period of forty years [i. e. since the early 1880s] are unaware of any examples being taken in

South Australia proper.” The last known specimen from Queensland was collected in 1905

(Archer 1979); the last western quoll from Victoria was collected in 1857 (Orell and Morris

1994), and the last specimen from the arid zone was in mid-1950s (Finlayson 1961).

20 Despite the dearth of records, Finlayson considered it to be relatively common throughout its range (Finlayson 1961).

The western quoll, like its congeners, falls within the critical weight range (35g-5500g) that is thought to be vulnerable to competition with (or predation by) dingoes, foxes, and cats . It is also a desert-dwelling species and populations would have been subject to great ecological stresses during the severe droughts earlier this century. However, this species is the most arid-adapted member of the genus and appears to have survived these pressures, surviving in some areas until the late 1950s (Finlayson 1961; Burbidge et al. 1988) even though the arid zone has suffered the most in terms of loss of species since European settlement (Burbidge and McKenzie 1989; Morton 1990). It has been suggested that a combination of factors including altered fire regimes, and intense de-vegetation by rabbits, in addition to competition with foxes, cats, and dingoes as causes for the decline of western quolls (Johnson and Roff

1982). Alteration of fire regimes, through the loss of traditional Aboriginal culture, indirectly may have caused a disruption to the availability of food resources through the lack of heterogeneity in the habitat.

Western quolls are currently restricted to a small remnant of their former range in southwest

Western Australia (Figure 4C) where they are thought to number approximately 6000 individuals (Orell and Morris 1994). They are being managed actively by the Western

Australian Department of Conservation and Land Management and Perth Zoo. Western quolls are considered endangered or extinct in most States where they once occurred and by

21 the Federal Government (Table 1). They are considered to be vulnerable by the World

Conservation Union (IUCN; Maxwell et al. 1996).

Northern quoll, Dasyurus hallucatus

Until recently, northern quolls were thought to be secure. However, like its congeners, this species has also suffered major declines in population numbers and distribution (Braithwaite and Griffiths 1994). The range of the northern quoll has decreased by 75% this century, and it is now reduced to several geographically isolated populations (Braithwaite and Griffiths

1994). These populations are restricted to coastal areas along the eastern, northern, and western coast of the continent (Figure 4D). Although this species has declined to a very great extent, the legal protection afforded to northern quolls is the least of all the dasyures. This species is considered to be common in Queensland, at low risk in the Northern Territory, and is not listed in Western Australia (Table 1).

The decline of northern quolls may be attributed to a number of factors including habitat destruction from tree clearing, the alteration of fire regimes (reducing the habitat mosaic), and the introduction of exotic herbivores (which destroy the vegetation supporting the food chain).

Another factor which may have impacted on the decline of this species is the introduction and spread of cane toads, Bufo marinus, which are poisonous to predators (Braithwaite and

Griffiths 1994; M. Oakwood, pers. comm.). The presence of cane toads may be a serious problem for quolls in the northern part of the continent and may have further adverse effects for local populations of quolls as the toads spread (Burnett 1997). One further factor which may be implicated in the decline

22 of this species is the possibility of a disease epidemic, like the one that possibly affected the southern species of quolls (Braithwaite and Griffiths 1994). The incidence of

Toxoplasmosis, however, was found not to be a major contributor to the decline of northern quolls in Kakadu National Park (Oakwood 1997). The species is not used as a food source by Aborigines, and introduced predators do not appear to be a major factor in direct predation of the northern quoll since they do not appear to be taken as a major food item by any species (Braithwaite and Griffiths 1994). However, the indirect effects of competition for resources by introduced species have not been accounted for.

Conservation Biology

As we approach the new millennium, the conservation of species, habitats, ecosystems, and the genetic diversity inherent within these systems has become a widely recognised concern for the global community. While human population numbers soar, the resources we use are being consumed at exceedingly rapid rates with the result of widespread degradation of ecosystems and loss of biodiversity. Proper functioning of ecosystem processes depends upon an intricate web of biotic and abiotic interactions and is the life support system that all living things depend on. If this complex web is damaged it may still function; however, if the web is greatly damaged, the processes that we all depend on for sustainable coexistence may no longer function properly.

The current rate of species extinctions is thought to be around 100 to 1000 times that of background extinction levels; this loss may be as high as one species per day. Within

23 Australia, 1424 taxa currently are either presumed extinct, endangered, or vulnerable

(Endangered Species Protection Act 1992, Schedule 1). This figure includes at least nineteen species or subspecies of mammals that are now thought to be extinct. The number of mammalian extinctions within Australia in the last 200 years is the highest of anywhere in the world, and currently all four members of the genus Dasyurus are under threat to some degree

(Maxwell et al 1996). Ideally, we would learn from past losses and conserve the remaining fauna and flora, the ecosystems that they depend on, and the genetic diversity retained within them.

Molecular Genetics in Conservation

Extinction is due primarily to habitat loss, competition with introduced species, overexploitation, and pollution (Diamond 1986). However, small populations may be prone to extinction due to both environmental and genetic factors. Rare, vulnerable, and endangered species are comprised of small populations which are of great concern to conservation agencies and managers. Inbreeding and outbreeding depression, the accumulation of deleterious alleles, loss of genetic variability, fragmentation of populations and reduced migration, and taxonomic uncertainties are some of the genetic factors that may jeopardise small populations (Frankham 1995a).

Conservation managers need to consider genetic differentiation concepts such as subspeciation and conservation units before examining within species variability. In the following series of papers, I explore potential genetic contributions to the conservation management of quolls. In particular, I examine the utility of both nuclear and mitochondrial genetic loci as a means of

24 determining 1) the evolutionary relationships among all six quoll species, 2) the degree of differentiation and levels of genetic variability within and among populations of the four

Australian species of quolls, and 3) the different conservation units found within tiger quolls

(Dasyurus maculatus).

An examination of historic evolutionary relationships among the species of quolls through phylogenetic reconstruction may provide a means of determining unique evolutionary groups and, thereby, distinguishing important lineages for conservation. An assessment of genetic variability within these species and populations is necessary to determine both the baseline levels of variation for future monitoring and to determine which, if any, populations may be a good source for future captive breeding or translocation programs. The analysis of differentiation between populations may assist in determining units for conservation purposes, e.g. Evolutionarily Significant Units (ESUs) and Management Units (MUs).

Phylogeny of the Quolls

Since the early 1800s there have been numerous phylogenetic, systematic, and taxonomic revisions of the Dasyuridae, the genus Dasyurus, and the species within this genus. Archer reviewed eighteen different studies up until the early 1980s (Archer 1982). Since his work, there have been at least six more studies directly or indirectly relating to the dasyures. Three general factors are responsible for the great interest in this family. First, all dasyurids resemble the American didelphid marsupials in retaining polyprotodont dentition and didactyl hind feet (as opposed to the derived conditions of diprotodonty and syndactyly) and they are, therefore, thought to represent the most plesiomorphic or ancestral Australasian marsupial

25 fauna. Secondly, many new species have been described over recent years, especially the

New Guinean and fossil taxa; the phylogenetic status of these newly described taxa has been examined extensively due to their enigmatic or unexplored positions. Finally, as newer methods became available (such as molecular techniques, DNA sequencing, sophisticated computer algorithms), additional independent character sets have been examined, and uncertain phylogenies reviewed.

Classic dental and osteological characters (Archer 1976; Archer 1982; Kirsch and Archer

1982; Wroe and Mackness 1998) in combination with external morphological characters

(Van Dyck 1987) have provided the basis of much of the recent phylogenetic work. Newer molecular techniques including comparative serology (Kirsch 1977), isozyme electrophoresis

(Baverstock et al. 1982), albumin immunology (Baverstock et al. 1990), DNA-DNA hybridization (Kirsch et al. 1990), and DNA sequencing (Krajewski et al. 1993; Krajewski et al. 1994; Krajewski et al. 1997) have also been used to determine inter- and intra-familial relationships, as well as relationships within the genus. I review some of these latter studies in

Chapter 3.

While there have been numerous studies at higher taxonomic levels which directly or indirectly relate to quolls, the phylogenetic affinities among these species have always been problematic.

Additional character sets may help to clear up these uncertainties. While morphological data

(dental and basicranial characters) have been used extensively in the past, these data are removed by a number of levels from the actual genetic diversity present. Previously, molecular data have confused the issue by indicating that quolls are a polyphyletic group

26 (Krajewski et al. 1994). In this thesis, I use another highly variable molecular locus (the mitochondrial DNA control region) in an attempt to clarify the relationships among quolls as well as to examine intraspecific phylogeographical lineages for conservation purposes.

The hypotheses examined are: 1) that the phylogenetic relationships among quolls follows that of Krajewski et al. (1997), and 2) there is distinct phylogenetic partitioning within species corresponding to distinct geographic localities.

Genetic Differentiation, Subspeciation, Conservation Units, and the ‘Agony of

Choice’

Given the limited financial resources available to conserve the many species that are declining, threatened, or endangered, much debate has been focused on what we can afford to conserve and what species or units should be considered as high priorities for conservation (Vane-

Wright et al. 1991; Crozier 1992; Rojas 1992; Vogler and DeSalle 1994; Crozier 1997).

Determining the degree of genetic differentiation between populations may provide assistance in deciding what to conserve. It may be that two populations that are genetically different are worth conserving as two separate units, however if these two populations are not genetically divergent, only one population may be worth conserving. Determination of genetic differentiation between populations may also provide guidelines as to how to manage these populations, as either Evolutionarily Significant Units (ESUs) or Management Units (MUs).

Despite the many taxa that are currently being conserved at the subspecific level, the fluidity and uncertainty of what a subspecies actually is, is a large stumbling block in

27 conservation of threatened or endangered taxa. Either the existing taxonomy may not adequately reflect the actual genetic diversity, or species may be split into many artificial taxa.

In addition, given the limited amount of space and financial resources available for conservation both in situ and ex situ, the need to determine the conservation status of subspecific taxa has come into focus. The definition of species (and subspecies) and the

‘agony of choice’ is a major theme in the conservation literature (Cracraft 1983; Ryder 1986;

Vane-Wright et al. 1991; Crozier 1992; Rojas 1992; Crozier 1997). The way in which species and lower categories are defined often affects the conservation goals and management decisions of wildlife managers, administrators, organizations, and governments.

For instance, the disregard of taxonomy has proved disastrous for some populations and has profound management implications for others. Twenty-seven subspecies of leopards were thought to exist at one time, but by using current molecular techniques, it was possible to collapse this number to eight subspecies (Miththapala et al. 1996), which more accurately reflects the true biology of this species. Currently, there are five subspecies of tigers that are being preserved in zoos or in the wild. However, there is some indication that the subspecific status assigned to tigers may be a political, rather than a biological reality (Bowdoin et al.

1994). The perceived monotypy of tuataras contributed to the extinction of 25% of the populations of one subspecies of these endangered reptiles (Daugherty et al. 1990).

Similarly, the rediscovery of extant populations of the Lake Eacham rainbow fish (which had been misdiagnosed as a related species) has led to a downgraded revision as vulnerable rather than extinct in the wild (Zhu et al. 1998). At present in Australia, over thirty subspecies of

28 marsupials have management plans or are being managed actively (Hopkins et al. 1997), and it is possible that some of these subspecies might not warrant such attention.

The use of taxonomic studies in conservation requires careful attention to the reason for making taxonomic decisions and a formal process for making these decisions. The notion of subspecies as a valid unit for conservation is often hotly debated (Ryder 1986). The concept of ESUs for conservation is related to taxonomic concept of subspecies. ESUs were first proposed as an attempt to clarify the ex situ conservation and management aims for a number of large charismatic taxa (e.g. leopards, tigers, rhinos) in which a number of subspecies of undetermined genetic status were being managed actively (Ryder 1986). However, defining

ESUs has proved to be almost as difficult as defining subspecies (Crozier 1992; Vogler et al.

1993; Krajewski 1994). Crozier (1992) has suggested that populations, and not species should be the most appropriate units for conservation. Cracraft (1983) suggested that ESUs should be based on the possession of unique phenotypes (whether morphological, biochemical, behavioural, physiological, etc.) which would allow for differentiation of taxa, and thus units for conservation; Vogler and DeSalle (1994) state that ESUs should be used as a technical term for clusters of organisms that are evolutionarily distinct and hence deserve separate protection.

Moritz (1994) and Moritz et al. (1995) suggested that both nuclear and mitochondrial loci should be considered as this enables more than one genetic system to be examined; they defined ESUs as those units that were reciprocally monophyletic for mitochondrial DNA alleles and possessed significantly divergent allele frequencies at nuclear loci.

29 This is the definition that I use in this work. However, there are still no good operational guidelines as to how much mitochondrial divergence is enough to distinguish these units. For example, is reciprocal monophyly due to 0.1% sequence divergence enough? Is 10% sequence divergence enough? Or is monophyly per se (in addition to significant differences at nuclear allele frequencies) the only criterion on which we base the definition of separate units?

ESUs are units more appropriate for ensuring the long-term evolution of taxa (Moritz et al.

1995) and are of use in conservation in that one would not mix stocks from two separate

ESUs since they are phylogenetically divergent.

In contrast to ESUs, defining MUs is more applicable to short-term conservation goals such as population monitoring and translocations. MUs have been defined as “populations with significant divergence of allele frequencies at nuclear or mitochondrial loci, regardless of the phylogenetic distinctiveness of the alleles” (Moritz 1994). MUs represent populations that currently are limited in genetic exchange, although exchanges between these units within one

ESU may be considered for translocation or supplementation programs in the case where population numbers have decreased to low numbers (Moritz 1999). However, as with determining ESUs, there are operational difficulties in determining MUs. For example, if two populations are significantly different at one locus out of ten, do they qualify as separate MUs?

Or do these two populations need to be significantly divergent at all loci to be considered as separate units?

30 Although genetic management is a priority in any modern recovery plan, the degree of differentiation or evolutionary uniqueness of any population of any species of quoll is unknown.

There have been no intraspecific phylogeographic studies of any of the dasyures. This may pose problems particularly where reintroductions may cause mixing of stocks. Currently, western quolls from a mesic habitat are being reintroduced into a xeric habitat within their former range, to establish a second population as a reserve against the possibility of stochastic disasters. For the same reason, there are also plans for reintroductions of eastern quolls into their former range on the mainland. The rapid rate of decline of the northern subspecies of tiger quoll, as well as that of the northern quoll, may eventually necessitate conservation action.

It is possible that mixing of stocks may occur where quolls are being reintroduced if remnant populations exist in these areas. Historic geographical isolation of the Tasmanian and north

Queensland populations of tiger quolls would suggest that these populations are possibly separate conservation units from those populations in the southeast of the mainland. Similarly, isolation of the Tasmanian eastern quoll populations from those formerly found on the mainland would suggest the possibility of separate conservation units. As stated earlier, northern quolls were once designated as four different subspecies (although the subspecific labels are no longer applied to this species); they are now confined to six isolated populations. In this thesis

I aim to clarify the degree of differentiation both between species and between populations within species of quolls, to effectively assist managers in conserving the genetic diversity of these species.

31 The null hypotheses examined are: 1) there is no genetic differentiation between any species of quolls, and 2) there is no genetic differentiation between any population within these species.

Genetic Variability

Genetic variation is fundamental to the well-being of populations, in that low levels of genetic diversity may have a negative influence on the ability of populations to adapt to changing environmental circumstances and may lead to extinctions (Frankel and Soulé 1981; Frankham

1985b; Lacy 1997). Genetic drift may cause the loss of genetic variability in small populations, however this may be ameliorated by immigration. Inbreeding may decrease genetic variability in individuals (Lacy 1987; Ralls et al. 1988). Individuals that have a low level of genetic variability may exhibit decreased reproductive fitness (Ralls and Ballou 1983;

Miller and Hedrick 1993), decreased resistance to disease (O'Brien et al. 1985; O'Brien and

Everman 1988), and decreased flexibility in response to environmental changes. For populations, low levels of genetic variation may lead to decreased mean fitness and decreased long-term adaptability (Lacy 1997).

Determining genetic variability is a key component of conservation genetics. Without sound knowledge of the amount of genetic diversity present within and between populations, it is difficult to determine which populations may be good sources for translocation and breeding programs. Where populations have been brought into captivity from the wild, assessment of genetic variability may assist in determining levels of inbreeding or founder effects.

32 It is important to assess the variability present within species to have a baseline measure from which to further assess populations. This baseline genetic variability is important as an indicator of overall population adaptability in the face of changing environmental circumstances. The greater the genetic variation, the more capable a population may be to adapt and evolve in the face of altered conditions. Low levels of variation may indicate population bottlenecks in the species history or the effect of inbreeding or founder effects.

However, the amount of genetic variation in quolls has never been examined. Pressing management issues for western quolls include assessment of the genetic diversity of the captive stock used for breeding and reintroductions. These issues necessitate examining the genetic diversity present within that species in particular. Among eastern quolls there is no available information regarding the degree of genetic variability within and among the remaining populations within Tasmania, yet there are continued proposals to reintroduce this species to the mainland. Similarly, among northern quolls and tiger quolls there is no information regarding the levels of genetic variability among populations within these species, yet with continued declines, these species will require management action. To provide this baseline information, this study has investigated all Australian species of quolls.

The null hypotheses examined are: 1) there is no difference in the amount of genetic variability between the different species of quolls and 2) there is no difference in the amount of genetic variability between different populations within species.

33 Objectives and Thesis Aims

The major objective of this study was to use molecular data to aid in the conservation and management of the large carnivorous marsupials of the genus Dasyurus. I aimed to do this by examining the evolutionary relationships of all members of the genus Dasyurus, by examining the levels of genetic variability and differentiation among and within populations of the four Australian species, and by establishing the conservation units and subspecific status of within tiger quolls.

Specifically, I have aimed to:

· construct a genomic library for the isolation and characterization of microsatellite loci

particular to quolls, as a first step in determining levels of variability and differentiation

among these species,

· provide a phylogenetic reconstruction using highly variable mtDNA control region

sequence data as a means of determining historical relationships among all species of

quolls,

· examine the levels of population subdivision, heterozygosity, and allelic diversity in the

Australian quolls using nuclear microsatellites, to provide both baseline levels of variability

within species and populations and provide a preliminary means of determining

conservation units,

· investigate the subspecific status and presence of ESUs and MUs among populations of

tiger quolls for conservation purposes using both the mitochondrial and nuclear loci,

34 · use the results of all these molecular analyses to guide management for the long-term

conservation of quolls in Australia.

References

Aplin KP, Archer M (1987) Recent advances in marsupial systematics with a new syncretic classification. In: Possums and Opossums: Studies in Evolution. (ed. Archer M) pp. xv- lxxii. Surrey Beatty and Sons and the Royal Zoological Society of New South Wales,

Sydney.

Archer M (1976) The Dasyurid dentition and its relationships to that of Didelphids,

Thylacinids, Borhyaenids (Marsupicarnivora) and Peramelids (Peramelina:Marsupialia).

Australian Journal of Zoology, Suppl. Ser. 39, 1-34.

Archer M (1978) Quarternary vertebrate faunas from the Texas Caves of southeastern

Queensland. Memoirs of the Queensland Museum, 19, 61-109, pls. 7-10.

Archer M (1979) The status of Australian dasyurids, thylacinids and myrmecobiids. In: The

Status of Endangered Australian Wildlife. (ed. Tyler, MJ) pp. 29-44. Royal Zoological

Society of South Australia, Adelaide.

Archer M (1982) Review of the dasyurid (Marsupialia) fossil record, integration of data bearing on phylogenetic interpretation, and suprageneric classification. In: Carnivorous

Marsupials (ed. Archer M) pp. 397-443. Royal Zoological Society of New South Wales,

Sydney.

35 Banks J (1896) Journal of the Right Hon. Sir Joseph Banks During Captain Cook's

First Voyage in H.M.S. Endeavour in 1768-71 to Terra Del Fuego, Otahite, New

Zealand, Australia, the Dutch East Indies, etc. Macmillan and Co. Ltd., London.

Bartholomai A (1971) Dasyurus dunmalli, a new species of fossil marsupial (Dasyuridae) in the Upper Cainozoic deposits of Queensland. Memoirs of the Queensland Museum, 16,

19-26, pl. 1.

Batey I (1907) The animal-life of the Sunbury district sixty years ago. Victorian Naturalist,

24, 69-74.

Baverstock PR, Archer M, Adams M, Richardson BJ (1982) Genetic relationships among

32 species of Australian dasyurid marsupials. In: Carnivorous Marsupials (ed. Archer, M) pp. 641-650. Royal Zoological Society of New South Wales, Sydney.

Baverstock PR, Krieg M, Birrell J (1990) Evolutionary relationships of Australian marsupials as assessed by albumin immunology. Australian Journal of Zoology, 37, 273-287.

Bowdoin J, Wiese RJ, Willis K, Hutchins M, Eds. (1994) AZA Annual Report on

Conservation and Science. American Association of Zoological Parks and Aquariums,

Bethesda, MD.

36 Braithwaite RW, Griffiths AD (1994) Demographic variation and range contraction in the northern quoll, Dasyurus hallucatus (Marsupialia:Dasyuridae). Wildlife Research, 21, 203-

217.

Burbidge AA, Johnson KA, Fuller PJ, Southgate RI (1988) Aboriginal knowledge of the mammals of the central deserts of Australia. Australian Wildlife Research, 15, 9-39.

Burbidge AA, McKenzie NL (1989) Patterns in the modern decline of Western Australia's vertebrate fauna: causes and conservation implications. Biological Conservation, 50, 143-

198.

Burnett S (1993) The conservation status of the tiger quoll, Dasyurus maculatus gracilis in north Queensland. Queensland Department of Environment and Heritage.

Burnett S (1997) Colonizing cane toads cause population declines in native predators: reliable anecdotal information and management implications. Pacific Conservation Biology, 3, 65-

72.

Caughley J (1980) Native quolls and tiger quolls. In: Endangered Animals of New South

Wales (ed. Haigh, C) pp. 45-48. National Parks and Wildlife Service, Sydney.

Cracraft J (1983) Species concepts and speciation analysis. In: Current Ornithology pp.

159-187. Plenum Press, New York.

37 Crozier RH (1992) Genetic diversity and the agony of choice. Biological Conservation, 61,

11-15.

Crozier RH (1997) Preserving the information content of species: genetic diversity, phylogeny, and conservation worth. Annual Review of Ecology and Systematics, 28, 243-

268.

Daugherty CH, Cree A, Hay JM, Thompson MB (1990) Neglected taxonomy and continuing extinctions of tuatara (Sphenodon). Nature, 347, 177-179.

Diamond J (1986) Design of a nature reserve system. In: Conservation Biology: The

Science of Scarcity and Diversity. (ed. Soulé, ME) pp. 485-503. Sinauer Associates,

Sunderland, MA.

Dickman CR, Read DG (1992) The Biology and Management of Dasyurids of the Arid

Zone in NSW. Species management report no. 11. NSW National Parks and Wildlife

Service, Hurstville.

Edgar R, Belcher C (1995) Spotted-tailed quoll. In: The Mammals of Australia (ed.

Strahan, R) pp. 67-69. Reed Books, Chatswood.

Finlayson HH (1961) On central Australian mammals. Records of the South Australian

Museum, 14, 141-191.

38 Firestone K (1999) Isolation and characterization of microsatellites from carnivorous marsupials (Dasyuridae:Marsupialia). Molecular Ecology, 8, 1084-1086.

Firestone K, Elphinstone M, Sherwin B, Houlden B (1999) Phylogeographical population structure of tiger quolls Dasyurus maculatus (Dasyuridae:Marsupialia), an endangered carnivorous marsupial. Molecular Ecology, 8, 1613-1626.

Firestone KB (in press) Phylogenetic relationships among quolls revisited: the mtDNA control region as a useful tool. Journal of Mammalian Evolution.

Firestone KB, Sherwin WB, Houlden BH, Geffen E (submitted) Variability and differentiation: microsatellites in the genus Dasyurus and conservation implications for the large Australian carnivorous marsupials. Conservation Genetics.

Flannery T (1995) Mammals of New Guinea. Reed Books, Sydney.

Fleay D (1932) The rare dasyures (native cats). Victorian Naturalist, 49, 63-68.

Fleay D (1948) Australia's marsupial tiger cat: a creature of increasing rarity and extreme pugnacity. Wildlife Australian Nature Magazine, 10, 459-464.

Fleay DH (1935) Breeding of Dasyurus viverrinus and general observations on the species.

Journal of Mammalogy, 16, 10-16.

39 Frankel OH, Soulé ME (1981) Conservation and Evolution. Cambridge University Press,

Cambridge.

Frankham R (1995a) Conservation genetics. Annual Review of Genetics, 29, 305-327.

Frankham R (1995b) Inbreeding and extinction: a threshold effect. Conservation Biology, 9,

792-799.

Geoffroy É (1803) Note sur les espèces du genre Dasyure. Bulletin des Sciences, par la

Society Philomatique de Paris, 3, 158-159.

Geoffroy É (1804) Memoire sur les espèces du genre Dasyure. Annales du Museum

D'Histoire Naturelle, Paris, 3, 353-363.

Geoffroy É (1797) Extrait d'une dissertation sur les animaux a bourses. Bulletin de la

Societe Philomatique de Paris, 1797, 106.

Gill T (1872) Arrangement of the families of mammals, with analytical tables. Smithsonian

Miscellaneous Collections. 40, i-iv and 1-98.

Godsell J (1995) Eastern quoll. In: The Mammals of Australia (ed. Strahan, R) pp. 70-71.

Reed Books, Chatswood,

Goldfuss GA (1820) Handbuch der Zoologie. JL Schrag, Nurnberg.

40 Gould J (1841) Observations of Dasyurus maugei and D. viverrinus of Geoffroy and description of a new species. Proceedings of the Zoological Society of London, 1840,

151.

Gould J (1842) Characters of a new species of Perameles, and a new species of Dasyurus.

Proceedings of the Zoological Society of London, 1842, 41-42.

Gray JE (1843) List of the Mammalia in the British Museum. The British Museum,

London.

Green RH (1967) Notes on the devil (Sarcophilus harrisii) and the quoll (Dasyurus viverrinus) in north-eastern Tasmania. Records of the Queen Victoria Museum, 40, 1-13.

Hampton JWF, Howard AE, Poynton J, Barnett JL (1982) Records of the mammal survey group of Victoria, 1966-80, on the distribution of terrestrial mammals in Victoria. Australian

Wildlife Research, 9, 177-201.

Hawkesworth J (1773) An Account of the Voyages Undertaken by the Order of His

Present Majesty for Making Discoveries in the Southern Hemisphere, and Successively

Performed by Commodore Byron, Captain Wallis, Captain Carteret, and Captain

Cook, in the Dolphin, the Swallow, and the Endeavour: Drawn up from the Journals

Which Were Kept by the Several Commanders, and from the Papers of Joseph Banks,

41 Esq. Illustrated with Cuts, and a Great Variety of Charts and Maps Relative to

Countries Now First Discovered, or Hitherto but Imperfectly Known. W. Strahan and T.

Cadell, London.

Hope JH (1972) Mammals of the Bass Strait Islands. Proceedings of the Royal Society of

Victoria, 85, 163-195.

Hopkins C, Johnson K, Wilcken J, Wilson C, Eds. (1997) Australasian Species

Management Program Regional Census and Plan. Australasian Regional Association of

Zoological Parks and Aquaria, Sydney.

Illiger C (1811) Prodromus systematis mammalium et avium additus terminis zoogeographicus utriudque classis. C. Salfeld, Berlin.

Iredale T, Troughton E (1934) A Check-list of the Mammals Recorded from Australia.

The Australian Museum, Sydney.

Johnson KA, Roff AD (1982) The western quoll, Dasyurus geoffroii (Dasyuridae:

Marsupialia) in the Northern Territory: historical records from venerable sources. In:

Carnivorous Marsupials. (ed. Archer M) pp. 221-226. Royal Zoological Society of New

South Wales, Sydney.

Jones ME (1995) Guild Structure of the Large Marsupial Carnivores in Tasmania. Ph D

Thesis. University of Tasmania, Hobart, Tasmania.

42 Kerr R (1792) The Animal Kingdom, or Zoological System, of the Celebrated Sir

Charles Linnaeus; Class I Mammalia. J. Murray and R. Faulder, London.

Kirsch JAW (1976) The classification of marsupials with a special reference to karyotypes and serum proteins. In: The Biology of Marsupials (ed. Hunsaker II, D) pp. 1-50.

Academic Press, New York.

Kirsch JAW (1977) The comparative serology of Marsupialia, and a classification of marsupials. Australian Journal of Zoology, Suppl. Ser. 52, 1-152.

Kirsch JAW, Archer M (1982) Polythetic cladistics, or, when parsimony's not enough: the relationships of carnivorous marsupials. In: Carnivorous Marsupials (ed. Archer M) pp.

595-619. Royal Zoological Society of New South Wales, Sydney.

Kirsch JAW, Calaby JH (1977) The species of living marsupials: an annotated list. In: The

Biology of Marsupials (eds. Stonehouse, B and Gilmore, D) pp. 9-26. Macmillan, London.

Kirsch JAW, Krajewski C, Springer MS, Archer M (1990) DNA-DNA hybridisation studies of carnivorous marsupials. II. Relationships among dasyurids (Marsupialia:

Dasyuridae). Australian Journal of Zoology, 38, 673-696.

Krajewski C (1994) Phylogenetic measures of biodiversity: a comparison and critique.

Biological Conservation, 69, 33-39.

43 Krajewski C, Painter J, Buckley L, Westerman M (1994) Phylogenetic structure of the marsupial family Dasyuridae based on cytochrome b DNA sequences. Journal of

Mammalian Evolution, 2, 25-35.

Krajewski C, Painter J, Driskell AC, Buckley L, Westerman M (1993) Molecular systematics of New Guinean dasyurids (Marsupialia:Dasyuridea). Science in New Guinea,

19, 157-166.

Krajewski C, Young J, Buckley L, Woolley P, Westerman M (1997) Reconstructing the evolutionary radiation of Dasyurine marsupials with cytochrome b, 12S rRNA, and protamine

P1 gene trees. Journal of Mammalian Evolution, 4, 217-236.

Lacépède BGE, (1799) Discours d'ouverture et de cloture du cours d'histoire naturelle...et tableaux methodiques des mammiferes et de oiseaux. Plassan, Paris.

Lacy RC (1987) Loss of genetic diversity from managed populations: interacting effects of drift, mutation, immigration, selection, and population subdivision. Conservation Biology, 1,

143-158.

Lacy RC (1997) Importance of genetic variation to the viability of mammalian populations.

Journal of Mammalogy, 78, 320-335.

44 Le Souef AS (1929) Notes on some mammals from Bass Strait Islands, including a new subspecies of Pseudochirus. Australian Zoologist, 5, 329-332 with one plate.

Mahoney JA (1964) The taxonomic status of Dasyurus affinis McCoy [1865] (Dasyuridae) and Hypsiprymnus trisulcatus McCoy [1865] (Macropodidae), two marsupials from a

Holocene cave deposit near Gisborne, Victoria. Proceedings of the Royal Society of

Victoria, 77, 525-533 with 2 plates.

Mahoney JA, Ride WDL (1984) The identity of Captain Cook's quoll Mustela quoll

Zimmermann, 1783 (Marsupialia:Dasyuridae). Australian Mammalogy, 7, 57-62.

Mahoney JA, Ride WDL (1988) Dasyuridae. In: Zoological Catalogue of Australia. Vol.

5. Mammalia. (ed. Walton, DW) pp. 14-33. Australian Government Printing Services,

Canberra.

Mansergh I (1984) The status, distribution and abundance of Dasyurus maculatus (tiger quoll) in Australia, with particular reference to Victoria. Australian Zoologist, 21, 109-122.

Marlow BJ (1958) A survey of the marsupials of New South Wales. CSIRO Wildlife

Research, 3, 71-114.

45 Marshall LG (1973) Fossil vertebrate faunas from the Lake Victoria region, S.W. New

South Wales, Australia. Memoirs of the National Museum of Victoria, 34, 151-171 with 3 plates.

Marshall LG, Case JA, Woodburne MO (1990) Phylogenetic relationships of the families of marsupials. In: Current Mammalogy (ed. Genoways, HH) pp. 433-505. Plenum Press,

New York.

Marshall LG, Hope JH (1973) A revaluation of Dasyurus bowlingi Spencer and

Kershaw 1910 (Marsupialia, Dasyuridae) from King Island, Bass Strait. Proceedings of the

Royal Society of Victoria, 85, 225-236.

Matschie P (1916) Die Verbreitung der Beuteltiere auf Neuguinea mit einigen Bemerkungen uber ihre Einteilung in Untergattungen. Mitteilungen aus dem Zoologischen Museum in

Berlin, 8, 259-308.

Maxwell S, Burbidge AA, Morris K, Eds. (1996) The 1996 Action Plan for Australian

Marsupials and Monotremes. Wildlife Australia, Canberra.

McCoy F (1865) Geological Survey of Victoria. Quarter Sheet 7 NW.

Miller PS, Hedrick PW (1993) Inbreeding and fitness in captive populations: lessons from

Drosophila. Zoo Biology, 12, 333-351.

46 Milne-Edwards MA (1880) On a new species of Dasyurus from New Guinea. The Annals and Magazine of Natural History; Zoology, Botany, and Geology, 5, 171-172.

Miththapala S, Seidensticker J, O'Brien SJ (1996) Phylogeographic subspecies recognition in leopards (Panthera pardus): molecular genetic variation. Conservation Biology, 10, 1115-

1132.

Moritz C (1994) Defining 'Evolutionarily Significant Units' for conservation. Trends in

Ecology and Evolution, 9, 373-375.

Moritz C, Lavery S, Slade R (1995) Using allele frequency and phylogeny to define units for conservation and management. American Fisheries Society Symposium, 17, 249-262.

Moritz, C (1999) Conservation units and translocations: strategies for conserving evolutionary processes. Hereditas, 130, 217-228.

Morrison PC (1945) Along the track with the editor. Wildlife Australian Nature Magazine,

7, 280-282.

Morton SR (1990) The impact of European settlement on the vertebrate animals of arid

Australia: a conceptual model. Proceedings of the Ecological Society of Australia, 16,

201-213.

47 Oakwood M (1997) The Ecology of the Northern Quoll, Dasyurus hallucatus. Ph D

Thesis. The Australian National University, Canberra.

O'Brien SJ, Everman JF (1988) Interactive influence of infectious disease and genetic diversity in natural populations. Trends in Ecology and Evolution, 3, 254-259.

O'Brien SJ, Roelke ME, Marker L, Newman A, Winkler CA, Meltzer D, Colly L, Evermann

JF, Bush M, Wildt DE (1985) Genetic basis for species vulnerability in the cheetah. Science,

227, 1428-1434.

Orell P, Morris K (1994) Chuditch recovery plan 1992-2001. Department of Conservation and Land Management, Como, WA.

Owen R (1877) Researches on the Fossil Remains of the Extinct Mammals of Australia: with a Notice of Extinct Marsupials of England. J. Erxleben, London.

Parker SA (1973) An annotated checklist of the native land mammals of the Northern

Territory. Records of the South Australian Museum, 16, 1-57.

Parkinson S (1773) A Journal of a Voyage to the South Seas, in His Majesty's Ship, the

Endeavour. Faithfully Transcribed from the Papers of the Late Sydney Parkinson,

Draughtsman to Joseph Banks, Esq. of his Late Expedition, with Dr. Solander, Round

48 the World. Embellished with Views and Designs, Delineated by the Author, and

Engraved by Capital Artists. S. Parkinson, London.

Phillip A (1789) The Voyage of Governor Phillip to Botany Bay; With an Account of the

Establishment of the Colonies of Port Jackson and Norfolk Island; Compiled from

Authentic Papers, Which Have Been Obtained from the Several Departments. To

Which are Added, the Journals of Lieuts. Shortland, Watts, Ball and Capt. Marshall;

With an Account of Their New Discoveries Embellished With Fifty Five Copper Plates, the Maps and Charts Taken from Actual Surveys and the Plans and Views Drawn on the Spot, by Capt. Hunter, Lieuts. Shortland, Watts, Dawes, Bradley, Capt. Marshall,

&c. J. Stockdale, London.

Pocock RI (1926) The external characters of Thylacinus, Sarcophilus, and some related marsupials. Proceedings of the Zoological Society of London, 1926, 1037-1084.

Ralls K, Ballou J (1983) Extinction: lessons from zoos. In: Genetics and Conservation (ed.

Schoenwald-Cox CM, Chambers SM, MacBryde B, Thomas WL) pp. 164-184. Benjamin

Cummings Publishing, Menlo Park, CA.

Ralls K, Ballou J, Templeton A (1988) Estimates of lethal equivalents and the cost of inbreeding in mammals. Conservation Biology, 2, 185-193.

Ramsay EP (1888) Notes on the fauna of the Bellenden-Ker Ranges. Proceedings of the

Linnean Society of New South Wales, 2, 1295-1299.

49 Ride WDL (1964) A review of the Australian fossil marsupials. Journal of the Royal

Society of Western Australia, 47, 97-131.

Ride WDL (1970) A Guide to the Native Mammals of Australia. Oxford University

Press, Melbourne.

Robinson NH (1987) Mammals of the national parks and nature reserves between Port

Hacking and the Shoalhaven River, Illawarra Heritage Committee.

Rojas M (1992) The species problem and conservation: what are we protecting?

Conservation Biology, 6, 170-178.

Ryder OA (1986) Species conservation and systematics: the dilemma of subspecies. Trends in Ecology and Evolution, 1, 9-10.

Schlegel H (1880) On Dasyurus albopunctatus. Notes From the Leyden Museum, 2, 51-

53.

Serena M, Soderquist T (1995) Western quoll. In: The Mammals of Australia (ed. Strahan,

R) pp. 62-64. Reed Books, Chatswood.

Shaw G (1800) General Zoology or Systematic Natural History. G. Kearsley, London.

50 Shepherd NC, Mahood IT (1978) The potential effect of feral dogs and cats on Australian native fauna. In: Australian Advances in Veterinary Science. pp. 108. The Australian

Veterinary Association, Sydney.

Spencer B, Kershaw JA (1910) A collection of sub-fossil bird and marsupial remains from

King Island, Bass Strait. Memoirs of the National Museum, Melbourne, 3, 5-35 with 8 plates.

Strahan R, Ed. (1983) The Australian Museum Complete Book of Australian Mammals.

Angus and Robertson, Sydney.

Strahan R, Ed. (1998) The Mammals of Australia. New Holland Publishers, Sydney.

Szalay FS (1982) A new appraisal of marsupial phylogeny and classification. In:

Carnivorous Marsupials. (ed. Archer M) pp. 621-640. Royal Zoological Society of New

South Wales, Sydney.

Tate GHH (1947) Results of the Archbold expeditions. No. 56. On the anatomy and classifications of the Dasyuridae (Marsupialia). Bulletin of the American Museum of

Natural History, 88, 97-156.

Tate GHH (1952) Results of the Archbold expeditions. No. 66. Mammals of Cape York

Peninsula, with notes on the occurrence of rain forest in Queensland. Bulletin of the

American Museum of Natural History, 98, 563-616.

51 Thomas O (1904) On some mammals from British New Guinea. The Annals and Magazine of Natural History; Zoology, Botany, and Geology, 14, 402-403.

Thomas O (1906) On mammals collected in south-west Australia for Mr. W.E. Balston.

Proceedings of the General Meetings for Scientific Business of the Zoological Society of London, 1906, 468-478.

Thomas O (1909) Some mammals from N.E. Kimberley, Northern Australia. The Annals and Magazine of Natural History; Zoology, Botany, and Geology, 8, 149-152.

Thomas O (1926) The local races of Dasyurus hallucatus. The Annals and Magazine of

Natural History; Zoology, Botany, and Geology, 9, 543-544.

Van Dyck S (1987) The bronze quoll, Dasyurus spartacus (Marsupialia:Dasyuridae), a new species from the savannahs of Papua New Guinea. Australian Mammalogy, 11, 145-156.

Vane-Wright RI, Humphries CJ, Williams PH (1991) What to protect? Systematics and the agony of choice. Biological Conservation, 55, 235-254.

Vogler AP, DeSalle R (1994) Diagnosing units of conservation management. Conservation

Biology, 8, 354-363.

52 Vogler AP, Knisley CB, Glueck SB, Hill JM, DeSalle R (1993) Using molecular and ecological data to diagnose endangered populations of the puritan tiger beetle Cicindela puritana. Molecular Ecology, 2, 375-383.

Waterhouse GR (1846) A Natural History of the Mammalia. H. Bailliere, London.

Watt A (1993) Conservation status and draft management plan for Dasyurus maculatus and

D. hallucatus in southern Queensland. Queensland Department of Environment and Heritage and the Department of Environment, Sport and Territories.

White J (1790) Journal of a Voyage to New South Wales with 65 Plates of Nondescript

Animals, Birds, Lizards, Serpents, Curious Cones of Trees and Other Natural

Productions. J. Debrett, London.

Wood-Jones F (1923) The Mammals of South Australia, Vol I-III. South Australian

Branch of British Science Guilds, Adelaide.

Wroe S, Mackness BS (1998) Revision of the Pliocene dasyurid, Dasyurus dunmalli,

(Dasyuridae; Marsupialia). Memoirs of the Queensland Museum, 42, 605-612.

Wroe, S (1996) Muribacinus gadiyuli, (Thylacinidae:Marsupialia), a very plesiomorphic thylacinid from the Miocene of Riversleigh, northwestern Queensland, and the problem of paraphyly for the Dasyuridae (Marsupialia). Journal of Paleontology. 70, 1032-1044

53 Wroe, S (1997) A reexamination of proposed morphology-based synapomorphies for the families of Dasyuromorphia (Marsupialia). I. Dasyuridae. Journal of Mammalian

Evolution. 4, 19-52

Wroe, S (1999) The geologically oldest dasyurid, from Miocene deposits of Riversleigh, northwestern Australia. Palaeontology. 42, 501-527.

Zhu D, Degnan S, Moritz C (1998) Evolutionary distinctiveness and status of the endangered

Lake Eacham rainbowfish (Melanotaenia eachamensis). Conservation Biology, 12, 80-

93.

Zimmermann EAW (1777) Specimen Zoologiae Geographicae, Quadrupedem Domicilia et Migrationes Sistems. Theodorum Haak, Lugduni Batavorum.

54 Chapter 2. Isolation and Characterization of Microsatellites from

Carnivorous Marsupials (Dasyuridae:Marsupialia)

Introduction

Quolls (Dasyurus spp.) are among the largest remaining carnivorous marsupials in Australia and New Guinea; as such, they occupy an important ecological niche. All six species of quolls are threatened to some degree (Flannery 1995; Maxwell et al. 1996), but the reasons for the observed declines in each species are varied and poorly understood at present. Management for conservation purposes is necessary to prevent further declines or loss of these species, however the lack of knowledge of the genetic variation and differentiation within members of this genus poses a serious stumbling block to effective conservation measures.

Microsatellites have a broad range of applications in conservation and evolutionary genetics and are widely used in population, parentage, and phylogeographic studies (Bruford et al.

1996). As a first step in the genetic management of quolls, six highly polymorphic microsatellite loci have been isolated and characterized according to the methods of Rassmann et al. (1991) and are described here.

Materials and Methods

Microsatellites were isolated from two partial genomic libraries. The first library was developed from the DNA of a Tasmanian eastern quoll (Dasyurus viverrinus). Approximately 20 mg DNA was restricted to completion with AluI and HaeIII. The resultant fragments were electrophoresed on a 2.0% agarose gel, and fragments approximately 80-250 bp in length were cut out, purified, and ligated into a pBS K/S+

55 vector at the SmaI site. Competent DH5a E. coli cells were transformed with the ligation products and plated onto selective Luria-broth media with ampicillin, X-Gal, and IPTG (LB- AXI plates; Sambrook et al. 1989) and grown overnight.

The second library was derived from approximately 100 mg of DNA pooled from a

Tasmanian eastern quoll (Dasyurus viverrinus) and a mainland tiger quoll (D. maculatus maculatus) restricted to completion with EcoRV, HpaI, SmaI, and BsuRI. The restricted

DNA was run on a 2.0% agarose gel and fragments approximately 180-400 bp in length were cut out and purified. The fragments were ligated into pCR-Script SK+ cloning vector restricted with the Srf I restriction enzyme (Stratagene). Epicurian-coli XLI-Blue MRF’ Kan supercompetent E. coli cells (Stratagene) were transformed and plated onto LB-MAXI agar plates (equivaltent to LB-AXI plates with the addition of 1 x methicillin) and grown overnight.

Colonies from both libraries were blotted onto Hybond-N nitrocellulose membranes

(Amersham) and fixed according to manufacturer’s recommendations. The membranes were pre-hybridized for two hours in a solution of 50% 6x SSC/0.05x Blotto (5% non fat dried milk, 0.02% sodium azide) and 50% formamide and hybridized overnight at 42 °C with a

32 [g- P]ATP labelled synthetic (CA)200-500 probe. Membranes were washed twice in 2x

SSC/0.1% SDS at room temperature for 5 min and once in 1x SSC/0.1% SDS at 68 °C for

1.5 hr. Membranes were exposed to X-ray film for up to one week at -70 °C.

56 Sixty-eight hybridizing colonies were isolated and cultured in LB broth; plasmid DNA was obtained by the TENS miniprep method (Zhou et al. 1990). The clones were sequenced in both the forward and reverse directions using the dideoxy chain termination method (Sanger et al. 1977) and Sequenase 2 enzyme (USB). Twenty-nine clones had microsatellites present.

Primer sites were identified for thirteen of these loci using PRIMER 0.5 (Lincoln et al. 1991).

Optimal polymerase chain reaction (PCR) conditions varied for each set of primers; 6 of these primer pairs produced unambiguous and polymorphic amplification products (Table 2).

Optimization PCRs (non-radioactively labelled reactions) were performed over a range of

annealing temperatures and MgCl2 concentrations. PCRs were carried out using approximately 50 ng of DNA, 0.5 mM of each primer, 0.5 - 8.0 mM MgCl2, 0.3 units of Tth

DNA polymerase (Biotech), 67mM Tris HCl pH 8.8, 16.6 mM (NH4)2SO4, 0.45% Triton X

100, 0.2 mg/mL gelatin, and 0.2 mM of each dNTP in a total volume of 10 mL. Once optimal

PCR conditions were determined (Table 2), radioactively labelled PCRs were performed to screen populations for variation. Radioactive PCR conditions were identical to optimized conditions, except that 0.08 mM forward primer end-labelled with [a-33P] ATP using T4PNK

(Promega) and 0.08 mM unlabelled reverse primer was used in a 10 mL reaction. Initial denaturation for 2 min at 92 °C was followed by 35-40 cycles of 0.5 min at 92 °C, 0.5 min at the optimal annealing temperature (Table 2), and 1 min at 72 °C with a final cycle of 10 min at

72 °C. Radioactive PCR products were resolved on 6% urea polyacrylamide gels with sequenced M13 as a size marker. PCR products were visualized after autoradiography for

24-72 hours.

57 Results and Discussion

Full microsatellite sequences are presented in Appendix 1 and all specimen voucher information and microsatellite scores are presented in Appendices 2 and 3 respectively. The five species of quolls analysed with these primers produced amplification products within a relatively narrow size range of each other. In addition, two primer pairs have produced polymorphic banding patterns in Antechinus minimus, a small carnivorous marsupial within the same family. All loci displayed a high number of alleles (Table 2). These primers are proving useful in elucidating many aspects of the conservation genetics of Dasyurus species

(Firestone et al. 1999; Firestone et al. submitted) and may prove useful for addressing similar questions in other closely related species of dasyurids.

58 15 14 23 19 17 19 No. of alleles (bp) 80-110 91-145 70-110 143-169 108-148 179-217 Size range . 270 253 276 263 293 248 examined Individuals 15 18 20 21 19 29 array (CA) (CA) (CA) (CA) (CA) (CA) Repeat Antechinus minimus Dasyurus species. ] 2 2 2.0 1.0 1.5 1.0 2.5 mM 0.625 [MgCl 1 ) ° 50 55 (C temp. 55-45 63-53 60-50 55-45 Annealing Amplifies successfully with 3 accession # AF124211 AF124212 AF124213 AF124214 AF124215 AF124216 GenBank 3’) ® Primer sequences (5’ R: TCATATAAGTCTTACTGTGCA F: AGGAAACTTCACAAGTGTCGA R: ATTAATGACTCATCTGTTGTTGG F: CAGCCCTTGAGTCTTGAGATT F: AATAGCAGAGACTCGATCC R: AGCCTTTATTACCTGGGAAG F: GAAATCCAAGCTCATTTTAG R: AATCAACTCTGGAATGCATC F: AATGCTAGATTTCACTCCC R: CCTCACATTTCTGGAACTG F: ATTGATGAACAAGACATAGCG R: CATACCACCCCAGGAGTTTC 3 3 PCRs were run at the higher temperature for first cycle and then in decreasing increments by 2 °C until lower Indicates final concentration in 10 ml reaction volume. Table 2. Six primer pairs used to characterize genetic variability and differentiation within Locus 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 1 was reached, then run for 35 to 40 cycles at the lower temperature or single indicated. 2

59 References

Bruford MW, Cheesman DJ, Coote T et al. (1996) Microsatellites and their applications to conservation genetics. In: Molecular Genetic Approaches in Conservation (ed. Smith TB,

Wayne RK) pp. 278-297. Oxford University Press, New York.

Firestone K, Elphinstone M, Sherwin B, Houlden B (1999) Phylogeographical population structure of tiger quolls Dasyurus maculatus (Dasyuridae:Marsupialia), an endangered carnivorous marsupial. Molecular Ecology, 8, 1613-1626.

Firestone KB, Sherwin WB, Houlden BH, Geffen E (submitted) Variability and differentiation: microsatellites in the genus Dasyurus and conservation implications for the large Australian carnivorous marsupials. Conservation Genetics.

Flannery T (1995) Mammals of New Guinea. Reed Books, Sydney.

Lincoln S, Daly M, Lander E (1991) PRIMER. Whitehead Institute, MIT.

Maxwell S, Burbidge AA, Morris K, Eds. (1996) The 1996 Action Plan for Australian

Marsupials and Monotremes. Wildlife Australia, Canberra.

Rassmann K, Schlotterer C, Tautz D (1991) Isolation of simple-sequence loci for use in polymerase chain reaction-based DNA fingerprinting. Electrophoresis, 12, 113-118.

60 Sambrook E, Fritsch F, Maniatis T (1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbour Laboratory Press, New York.

Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating inhibitors.

Proceedings of the National Academy of Sciences of the USA, 74, 5463-5467.

Zhou C, Yang Y, Jong AY (1990) Miniprep in ten minutes. BioTechniques, 8, 172-173.

Acknowledgements:

This research was funded by Sydney City Council, Cassegrain Wines, the Winifred Scott

Foundation, and Taronga Zoo. I thank P. England for assistance with manufacturing and screening the original genomic library. B. Houlden and W. Greville provided technical assistance and advice through various stages of this work. S. Burnett, G. Hall, M. Jones, D.

Moyle, M. Oakwood, and M. Westerman kindly provided tissue samples used in this study. I also thank D. Edwards who trialled these primers on Antechinus species.

61 Chapter 3. Phylogenetic Relationships Among Quolls Revisited: the mtDNA Control Region as a Useful Tool

There has been a great deal of interest in determining phylogenetic relationships within the family Dasyuridae due to the widespread distribution, ecological diversity, and relative plesiomorphy of this taxon within the Australasian marsupial radiation. In the past, it has been extremely problematic to determine the phylogenetic relationships among species within

Dasyurus, with numerous studies using both morphological and molecular characters providing different topologies. Here, the mitochondrial DNA (mtDNA) control region is used as a novel set of characters in an attempt to identify relationships among the six closely-related extant species. Sequences were obtained from multiple individuals representing all extant species of quolls including, when possible, individuals from different geographical regions.

Sequences were analyzed using both parsimony criteria and neighbor-joining methods.

Results presented here concur with those of Krajewski et al. (1997) in (1) placing D. geoffroii in a highly supported clade with D. spartacus, (2) resolving a monophyletic group of D. albopunctatus + D. geoffroii + D. spartacus, and (3) placing D. hallucatus as the sister taxon to all other species of quolls. Results also show two highly supported and geographically distinct clades of D. maculatus (Tasmanian and mainland) that do not correspond to the currently used subspecific nomenclature. Preliminary results also indicate that there are different clades among geographic groups of D. hallucatus that warrant further investigation. The mtDNA control region is a highly variable locus and may be used in forensic tests for species identification in this genus.

62 Introduction

The family Dasyuridae is a carnivorous/insectivorous marsupial taxon comprised of at least 60 species (Flannery, 1995; Strahan, 1998). The genus Dasyurus includes six extant species of medium to large carnivores known as quolls. The quolls are the dominant carnivorous marsupials on the mainland of Australia and in Papua New Guinea. Currently all four

Australian species are at risk, threatened, or endangered, but little is known of the status of the two New Guinean species, D. albopunctatus and D. spartacus (Flannery, 1995; Maxwell et al., 1996). The determination of evolutionary relationships among the dasyurids, and particularly within the genus Dasyurus, remains problematical despite numerous attempts to characterize these associations (e.g. Archer, 1976b; Baverstock et al., 1982; Van Dyck,

1987; Krajewski et al., 1994; Krajewski et al., 1997; Wroe and Mackness, 1998). The great interest in this group stems partially from the hypothesis that the dasyurids are the most plesiomorphic of the Australasian marsupial radiation (Kirsch, 1977; Archer, 1984; Marshall et al., 1990). There is, however, debate on whether or not the bandicoots

(Peramelidae:Marsupialia) represent one of the basal branches within Australidelphia (Kirsch et al., 1991; Springer et al., 1994).

Problems in determining the relationships among the dasyurids and, in particular, among the quolls arise from a number of factors including missing taxa, the numerous different character sets examined and methods of analysis employed, inclusion (or exclusion) of outgroups, and questions of homology, polarity, and independence of characters. In some previous phylogenetic reconstructions the New Guinean taxa were

63 not considered (e.g. Archer, 1976b; Baverstock et al., 1982) or no outgroup was included

(e.g. Van Dyck, 1987; Wroe and Mackness, 1998). In other works higher level relationships were focal and not all species of quolls were included in analyses (e.g. Kirsch et al., 1990a;

Krajewski et al., 1994).

In addition, various characters and analytical techniques have been used to determine the historical evolution of quolls. Earlier efforts include both morphological [dental (Archer,

1976b; Wroe and Mackness, 1998) and basicranial characters (Archer, 1976a; Van Dyck,

1987)] and molecular characters [albumin immunology (Baverstock et al., 1990), allozyme electrophoresis (Baverstock et al., 1982), DNA-DNA hybridization (Kirsch et al., 1990b),

DNA sequencing (Krajewski et al., 1994; Krajewski et al., 1997)], and each study produced a different topology.

Furthermore, another debate has focussed on the placement of the northern quoll. Opinion has vacillated over the years with some researchers placing the northern quoll in a separate genus Satanellus (Pocock, 1926; Archer, 1982; Kirsch and Archer, 1982; Marshall et al.,

1990) while others include it in Dasyurus (Baverstock et al., 1982; Van Dyck, 1987;

Baverstock et al., 1990; Kirsch et al., 1990a; Krajewski et al., 1994; Wroe and Mackness,

1998). In this paper I use the taxonomic decision of Mahoney and Ride (1984) in referring to this species as a member of Dasyurus.

Review of Past Phylogenetic Reconstructions of Quolls

A number of important studies have examined the phylogenetic relationships among the quolls; a brief chronological review of some of these studies follows (see also Krajewski

64 et al., 1997). Archer (1976b) presented a phenogram of similarity based on dental characters for a large group of masupials (Figure 5). He concluded that the dasyurids were a unified group and that the Tasmanian devil (Sarcophilus harrisii) was a derivative of the Dasyurus lineage. Of the extant Australian quolls, D. maculatus was hypothesized to have diverged from the other quolls first, while D. hallucatus was thought to have diverged later. D. geoffroii and D. viverrinus were shown to be sister species, and the most highly derived of the quolls. Neither of the New Guinean taxa were included in this study (D. spartacus had not yet been described).

Figure 5. Partial phenogram of dental similarity. Redrawn from Archer (1976b). D. = Dasyurus; S. = Sarcophilus. The fossil forms Glaucodon (a devil) and D. dunmalli (a quoll) were included in this study.

D. geoffroii

D. viverrinus

D. hallucatus D. maculatus

D. dunmalli Glaucodon

S. harrisii

In another comprehensive study of higher level phylogeny, serological techniques were used to classify approximately 100 species representing all families of both Australasian and American marsupials (Kirsch, 1977). While the focus of this study was on higher level relationships,

65 there was some resolution of relationships among the quolls. The finer evolutionary relationships should be interpreted with caution, however. This study largely concurred with

(Archer, 1976b) in the placement of D. geoffroii and D. viverrinus as sister taxa, and in placement of D. hallucatus intermediate between the D. geoffroii + D. viverrinus clade and

D. maculatus (Kirsch, 1977).

Allozyme electrophoresis has also been used to determine the genetic relationships of the dasyurids (Baverstock et al., 1982; Figure 6). While the focus was also on higher evolutionary relationships, the affinities among the quolls were resolved to some degree.

These authors found that, as in the above mentioned studies, D. geoffroii and D. viverrinus were sister taxa, but in this case they were monophyletic with respect to D. maculatus. The relationships of the geoffroii/viverrinus/maculatus clade to D. hallucatus and Sarcophilus harrisii were unresolved. Sarcophilus harrisii occurred within the Dasyurus clade using

Wagner analysis (phylogeny based on parsimony analysis), but not when using Hennig analysis

(phylogeny based on synapomorphies; Baverstock et al., 1982). The New Guinean species were not included in these analyses.

Van Dyck (1987) used dental, basicranial, and other morphological characters in a description of a new species, D. spartacus (Figure 7). In his reconstruction, D. hallucatus appeared to be the most plesiomorphic of the quolls. D. spartacus and D. albopunctatus were synapomorphic with regard to a crushed premolar toothrow and appeared to be related more closely to D. maculatus than to any other extant species. The New Guinean species appeared to be a late diverging lineage whereas D. hallucatus was an early diverging lineage.

D. dunmalli, a fossil species, formed a clade with D. spartacus and D. albopunctatus; these

66 three taxa were monophyletic with respect to D. maculatus. Van Dyke assumed monophyly of the quolls and no outgroup was included in the analysis.

Figure 6. Cladistic relationships of the Australian quolls based on allozyme electrophoresis. Redrawn from Baverstock et al. (1982).

D. hallucatus

D. geoffroii

D. viverrinus

D. maculatus

S. harrisii

Figure 7. Cladistic relationships of quolls based on dental, basicranial, and morphological characters. Redrawn from Van Dyck (1987).

D. albopunctatus D. spartacus

D. dunmalli

D. maculatus

D. geoffroii

D. viverrinus D. hallucatus

67 Partial cytochrome b sequences (657 bp) were used to examine dasyurid relationships

(Krajewski et al., 1993; Krajewski et al., 1994). Results from this analysis indicated that the quolls were a polyphyletic group. In this scenario, D. spartacus and D. viverrinus were sister species that formed a highly supported clade with D. albopunctatus; in addition with

Dasykaluta, this clade formed an outgroup to other members of the subfamily Dasyurinae. D. hallucatus was the sister species to a D. maculatus + Sarcophilus clade. However, other than the spartacus + viverrinus + albopunctatus clade, many of these branch nodes were only weakly supported by bootstrapping. This study did not include a critical taxon (D. geoffroii), and the authors declined to talk about these findings except to say that these results

“suggest several intriguing relationships at the level that we will not discuss in detail here”

(Krajewski et al., 1994, p.31).

Stimulated by the anomalous results from partial cytochrome b sequences, Krajewski et al.

(1997) provided another molecular reconstruction based on complete cytochrome b, protamine P1, and 12S rRNA sequences. The primary focus of this work was on generic relationships within the subfamily Dasyurinae, although species-level relationships among quolls were examined in detail. These results corroborate the monophyly of Dasyurus, and they support Van Dyck (1987) in placing D. hallucatus as a sister to all other quolls (Figure 8).

This reconstruction, however, places the New Guinean taxa in a clade with D. geoffroii, although only weakly supported by bootstrap analysis. D. viverrinus, D. geoffroii, D. albopunctatus, and D. spartacus form a monophyletic clade with regard to the other species of quolls. Most of the quoll nodes

68 in this reconstruction are moderately to highly supported by bootstrapping, as opposed to the low bootstrap support presented in Krajewski et al. (1994).

Figure 8. Majority rule consensus tree based on complete cytochrome b, 12S rRNA, and protamine P1 sequences. Redrawn from Krajewski et al. (1997). D. = Dasyurus; S. = Sarcophilus. D. geoffroii

D. spartacus

D. albopunctatus

D. viverrinus

D. maculatus D. hallucatus S. harrisii

Wroe and Mackness (1998) used molar morphology to reconsider the position of the fossil species, D. dunmalli within Dasyurus (Figure 9). Their results support Van Dyck (1987) and

Krajewski et al. (1997) in that the extant New Guinean quolls did not form a monophyletic group with D. hallucatus but were interpreted to be more highly derived than previously assumed. D. hallucatus is the most plesiomorphic species forming a sister to the other quolls;

D. geoffroii branches off next, then D. viverrinus. The position of the New Guinean taxa, D. albopunctatus and D. spartacus, relative to each other and to D. maculatus was not resolved. D. dunmalli was considered to be more

69 derived and the sister taxon to D. maculatus. Their hypothesized topology should be interpreted with caution and retested when further fossil material becomes available. In addition, the authors assumed monophyly of Dasyurus and did not include any outgroup taxa.

Figure 9. Phylogenetic relationships among Dasyurus based on molar morphology. Redrawn from Wroe and Mackness (1998). D. maculatus

D. dunmalli D. spartacus D. albopunctatus

D. viverrinus

D. geoffroii

D. hallucatus

The Control Region as a Phylogenetic Marker

As mentioned above, previous studies using molecular techniques (Krajewski et al., 1994;

Krajewski et al., 1997) have examined phylogenetic relationships at a higher level (among and within tribal lineages of the Dasyuridae) and have used slower evolving molecular markers more suitable for such purposes.

70 The mitochondrial DNA (mtDNA) control region was chosen as a marker in this study due to several characteristics: 1) mtDNA is predominantly maternally inherited and not subject to the effects of recombination, although there is some evidence of paternal leakage (Gyllensten et al., 1991); 2) the control region is a noncoding locus and is not under the same evolutionary constraints as coding genes; therefore it exhibits relatively high substitution rates and levels of variability; and 3) because of these attributes, the control region can provide fine-scale resolution among closely related species and within species. In addition, the control region has not been used previously in studies of dasyurid evolution.

In this paper, I expand on earlier molecular reconstructions (Baverstock et al., 1982;

Krajewski et al., 1994; Krajewski et al., 1997) using another set of characters (552 base pairs of the control region) in an attempt to more clearly resolve phylogenetic relationships within the genus Dasyurus. In addition, for the first time, the fine scale intraspecific relationships of quolls is examined as a means of determining distinct lineages for conservation purposes. In general, each character set employed previously has provided a unique topology, and no single hypothesis has been strongly supported. In this study, the phylogenetic reconstructions produced from control region sequences largely concur with the work of Krajewski et al. (1997).

Materials and Methods

Samples and DNA Extractions

Samples used in this study were collected as skin biopsies stored in 80% ethanol, whole blood in EDTA, or liver biopsies from road-killed animals stored frozen at -20° C.

71 Since the mtDNA control region is so variable, multiple individuals from each species were included in the analysis when possible (Wiens and Servedio, 1997). Also when available, samples from different geographic regions of the specie’s range were included. All specimens with different haplotypes that were used in this study are listed in Table 3.

Total genomic DNA was prepared by standard phenol-chloroform extraction followed by ethanol precipitation (Sambrook et al., 1989). The outgroups chosen were: a Tasmanian devil, Sarcophilus harrisii, as member of a different genus within the same family

(Dasyuridae); a numbat, Myrmecobius fasciatus, as a member of a different family within the same order (Dasyuromorphia); and a northern brown bandicoot, Isoodon macrourus, as a member of a different order within the same cohort (Australidelphia) (Aplin and Archer,

1987). Numbat and bandicoot sequences were provided by L. Fumagalli. The choice of outgroups is important for determining where the root of the tree is placed. However, the initial choice of outgroups was constrained by the limited number of species for which control region sequence was available; the numbat and bandicoot sequences were the closest related sequences to Dasyuridae that were obtainable.

72 GenBank AF082751 AF082761 AF082764 AF082765 AF082768 AF082763 AF082770 AF082757 AF082759 AF082753 AF082762 AF082754 AF082776 AF082777 AF082774 AF082775 AF082778 AF082779 AF082780 AF082771 AF082772 Accession # 4 1 N 1234 24 5 9 6 2 7 5 8 1 9 3 3 5 1 101112 45 13 3 14 7 1516 1 17 1 18 1 1 1 19 1 20 30 1 ID# Haplotype 3 ID# NM.B1 QM.E1 TM.E5 TM.E6 TM.E10 TM.E1 VM.L2 NM.E45 NM.L12 NM.E2 QM.E2 NM.E8 TV.E10 TV.E15 TV.E12 TV.E13 TV.E14 TV.E11 TV.M58 WG.E1 WG.E7 Individual 2 Source KF KF/S. Burnett, JCU M. Jones, UT M. Jones, UT M. Jones, UT M. Jones, UT KF/J. Seebeck C. Belcher, LU KF KF KF/S. Burnett, JCU KF KF/D. Moyle, UT M. Jones, UT M. Jones, UT M. Jones, UT M. Jones, UT M. Jones, UT KF/J. Rainbird, QVM G. Hall, Perth Zoo G. Hall, Perth Zoo Tissue Type whole blood skin biopsy skin biopsy skin biopsy skin biopsy skin biopsy liver skin biopsy liver skin biopsy skin biopsy skin biopsy skin biopsy skin biopsy skin biopsy skin biopsy skin biopsy skin biopsy museum skin skin biopsy skin biopsy 1 ML-N ML-Q TAS TAS TAS TAS ML-V ML-N ML-N ML-N ML-Q ML-N TAS TAS TAS TAS TAS TAS TAS WA-M WA-X Locality Code Species Table 3. Samples used in this study Dasyurus maculatus (tiger quoll) D. viverrinus (eastern quoll) D. geoffroii (western quoll)

73 Source: 2 GenBank AF082773 AF157095 AF082785 AF082788 AF082787 AF082781 AF082784 AF082783 AF082782 AF157094 Accession # 4 N 2122 1 23 1 242526 1 27 2 28 1 29 1 1 30 5 1 31 1 32 1 1 ID# The representative individual 3 Haplotype 3 ID# NGS.1 NGA.L1 NTH.L6 QH.E3 NTH.L10 NTH.L15 NTH.L4 NTH.L1 NTH.E8 SHAR.1 Individual 2 Source M. Westerman, LU S. Donellan, SAM M. Oakwood, ANU L. Pope, UQ M. Oakwood, ANU M. Oakwood, ANU M. Oakwood, ANU M. Oakwood, ANU M. Oakwood, ANU D. Colgan, AM L. Fumagalli, UQ L. Fumagalli, UQ Tissue Type DNA liver liver skin biopsy liver liver liver liver skin biopsy DNA sequence sequence 1 PNG PNG NT QLD NT NT NT NT NT TAS WA QLD Locality Code The number of individuals screened that had the same haplotype. 4 Species Locality codes: ML-N = mainland, New South Wales; ML-V Victoria; ML-Q Queensland; TAS Tasmania; WA-X Table 3. Continued D. spartacus (bronze quoll) D. albopunctatus (New Guinea quoll) D. hallucatus (northern quoll) Sarcophilus harrisii (Tasmanian devil) Myrmecobius fasciatus (numbat) Isoodon macrourus (brown bandicoot) 1 Western Australia, xeric; WA-M = mesic; NT Northern Territory; PNG Papua New Guinea; QLD Queensland. AM = Australian Museum; ANU National University; JCU James Cook LU Latrobe QVM Queen Victoria Museum; SAM = South Australian UQ University of Queensland; UT Tasmania. sequenced.

74 Control Region Amplification, Screening, and Sequencing

Partial sequences from the 5’ variable end of the control region were amplified by PCR using primers (mt15999L and mt16498H) developed to amplify across a wide range of marsupial species (Fumagalli et al. 1997). Amplification of this segment was achieved by adding ~20-

100 ng DNA, 2.5 mM MgCl2, 0.75 units of Tth Plus polymerase (Biotech International, Ltd.),

67 mM Tris-HCl pH 8.8, 16.6 mM (NH4)2SO4, 0.45% Triton X-100, 0.2 mg/mL gelatine,

0.1 mM each primer, and 0.2 mM each dNTP in a final volume of 25mL. Thermal cycling was performed by denaturing at 94°C for 2 min followed by 35cycles of denaturation (94°C, 45 sec), annealing (50°C, 45 sec), and extension (72°C, 1 min) with a final extension of 5 min at

72°C. A small aliquot of each PCR product was electrophoresed on a 2.0% agarose gel to verify amplification. Products were then precipitated using either WizardÔ PCR Preps

(Promega) or a polyethylene glycol precipitation, and electophoresed again on a 2.0% agarose gel to quantify the amount of DNA present.

Prior to sequencing, PCR amplification products were screened for haplotypic variants using temperature gradient gel electrophoresis (TGGE) (Campbell et al., 1995; Elphinstone and

Baverstock, 1997). For each variant identified by TGGE the control region was sequenced using the same primers used in PCR amplifications (Fumagalli et al., 1997). Control region

PCR products were sequenced using either the ABI PRISMÔ Dye Terminator or BigDyeÔ

Terminator Cycle Sequencing Ready Reaction kits with AmpliTaqÒ DNA polymerase FS.

Cycle sequencing was performed in a Perkin Elmer 9600 thermal cycler by denaturing at

96°C for 5 sec followed by 25 cycles of denaturation (96°C, 10 sec), annealing (50°C, 5 sec) and extension (60°C, 4 min). Sequencing was performed on an ABI 377 automated

75 sequencer. Each unique haplotype was sequenced in both directions. All unique sequences have been deposited in GenBank (Table 3).

Statistical Analyses

The unique control region sequences were aligned using the CLUSTAL W algorithm

(Thompson et al., 1994), and further adjusted manually to eliminate unnecessary gaps. The numbat sequence used here as an outgroup taxon, incorporated four copies of a large (77 bp) repeat segment. Three of these repeat units were removed from the sequence to facilitate alignment. Percent base composition and transition-transversion ratios were estimated using

MEGA (Kumar et al., 1993).

Phylogenetic relationships among the species were assessed by a number of methods using all positions in the alignment for these anlayses. Firstly, Tamura distances were generated by using pairwise-deletion of gaps from the sequence data (Tamura, 1992); these distances were used to build a neighbor-joining tree in MEGA (Kumar et al., 1993). Tamura distance was chosen because of the high A+T ratio in the sequences (75% AT content) and the bias towards transitions over transversions (1.43:1). The support for this branching pattern was assessed by 500 bootstrap replicates. Secondly, evolutionary relationships were assessed using the maximum parsimony algorithm as implemented in PAUP 4.0b2 (Swofford, 1999) with gaps treated as missing data. All minimum trees were saved using the heuristic search algorithm and a majority rule consensus tree was constructed. The support for this branching pattern also was assessed by 500 bootstrap replicates. The Kishino-Hasegawa maximum likelihood test

76 was performed to determine whether different trees are significantly different (Kishino and

Hasegawa, 1989). Tests were performed of the currently proposed topology from mtDNA against previously proposed topologies based on morphological data. Constraint trees were made in MACLADE 3.01 (Madison and Madison, 1992) and tested in PUZZLE 4.0.2

(Strimmer and von Haeseler, 1996). Thirdly, the relationships among these species were analysed using a minimum spanning network (Excoffier and Smouse, 1994) which provides a network of connecting haplotypes rather than a simple bifurcating tree. A pairwise matrix of the number of nucleotide substitutions was used as the input data for this analysis.

Results

Thirty different control region haplotypes were found among the different species of quolls

(Table 3, Figure 10). The complete alignment is shown in Appendix 4. Complete sequences were obtained for all but NM.E45, TV.E10, TV.E15, TV.E14, TV.E11 and TV.M58. The

TV.M58 sequence was not included in further analyses, due to a large number of missing characters. In this study only the 5’ variable portion of the control region, homologous to the

L-strand positions 15999-16498 in the human mtDNA genome (Anderson et al., 1981), was examined.

The final sequence length, including insertions for alignment purposes, was 552 bp. There were 299 (54.2%) variable sites of which 148 (26.8%) were parsimony informative (Figure

10). The mean overall nucleotide composition from this portion of the control region is 39.0%

A, 35.6% T, 17.3% C, and 8.1% G. Starting from the 5’ end of this region, the sequence found in quolls is characterized by a very variable and

77 Figure 10. Parsimony informative sites of 30 control region sequences quolls and the Tasmanian devil. Numbers along top indicate position within the aligned sequences of L-strand. Sequences are numbered as in Table 3. A (?) = missing data, (.) identity with top consensus sequence, (-) = inserted gaps, letters indicate substitutions. 111111111111111111111111111111111111111111111222222222222233333333333333333333333333444444444444444444444444444455555555 122233333444444555555667789111111333334444555555666666777777788888999999002334445889900000111111233333333444778001122233344444456667777899912222334 3345601279123478012347060673124579356781678046789023789012368924568123467042050792892401457245678512345679035788347907912901478950562367502360123026 C ACTTTCTTTTAATTTACACAATTAGTTTTTTTTCCAAATTAATCCAATTTTTATATATTCTATTTAATATAAAATAATAGCGTAAAGCTCTATCAATATAAACTATAACTAGGCTAAGAGAATAGAATTGCATAAATTCCCTCATACA 1 .------...... A...... -C....---A...... T.C--...C..-....G.C-.G...... T...... A....GA..G...... T...... T...... C...... TC...... T. 2 .------.....CA...... -C....---A...... T.C--...... -....G..-....C...T...... GAT...GA..G...... C...... TC...... 3 .------...... A...... -.G..A---...... T..--...C..-..T.G..-..C.C...T...... T...GA..G...... T...... A..CT...... TC...... 4 .------...... A...... -.G..A---A...... T.C--...C..-...... -..C.C...T...... T...GA..G...... T...... A..CT...... TC...... 5 .------...... A...... -.G..A---...... T..--...C..-....G..-..C.C...T...... T...GA..G...... T...... A..CT...... TC...... 6 .------...... A...... -.G..A---...... T.C--...C..-....G..-..C.C...T...... T...GA..G...... T...... A..CT...... TC...... 7 .------...... A...... -C....---A...... T.C--...C..-....G.C-.G..C...T...C...... A....GA..G...... G...... TCC...... 8 ?????????????????????-C....---A...... T.C--...C..-....G.C-.G..C...T...... A....GA..G...... T...... C...... TC...... 9 .------...... A...... -C....---A...... T.C--...C..-....G.C-.G...... T...... A....GA..G...... T...... T...... C...... TC...... T. 10 .------...... A...... -C....---A...... T.C--...C..-....G.C-.G..C...T...C...... A....GA..G...... C...... TC...... C... 11 .------...... A...... -C....---A...... T.C--...C..-....G.C-.G..C...T...... C...A....GA..G...... C...... TC...... 12 .------...... A...... -C....---A...... T.C--...... -....G..-....C...T...C...... G.T...GA..G...... T...... C...... TC...... 13 ????????..TTA....GA..-...... T...-.C..CT..C..-...... G...T.A...... A...T...... T...... CT...... T..G...... TC...... ?G.?T. 14 ??????????????????????????..-....T...-.C..?T..C..-...... G...T.A...... A...T...... T...... CT...... T..G...... TC...... ?G.?T. 15 .----..A..TTA.....A..-...... T...-.C..CT..C..-...... G...T.A...... A...T...... T...... CT...... T..G...... ?TC....A...... ?T. 16 C----..A..TTA.....A..-...... -....T...-.C..CT..C..-...... G...T.A...... A...T...... T...... CT...... T..G...... TTC....A...... ?T. 17 ????????????????????????????.....T...-.C..CT..C..-...... G..-T.A...... A...T...... T...... CT...... T..G...... TC....A...... G..T. 18 ????????????????????????????????????????????A....-?..-....AT...... G...T.A...?...... AAA..A...... T...... TA..?...... TC...... A.CTG?TA. 19 .----TA....T...TT....AC..C.-A....---...... T...A...... T..C...... T.A..C...... AA..T...T....T...... A...... TCAG...... T...CT.G 20 .----TA...T....TT....AC..C.-A....---...... T...C...... T..C.CT...... T.A...... AA..T...T...GT...... A...... TCAG...... T...CT.G 21 C----T...ATT...TT....AC..C.-AA...---...... T...A...... T..C.C...... T.A..C...... AA..T...T....T....T...... TC.G...... T....T.G 22 .TACAT...CTTA...T....-...... ------...... -...A.CAC.C..C.TC.C..T.....TC.A...... A.....G....T...... A...... TC...... TC...T.G 23 -...... C...-.TT-.....A..------TT...C.TT...--.-..C.T.A..T...... G....C....AC..A.ATA..AG...... C...GT.AA.T.A..T...GG-..TTAA..T.....CAA...... G 24 -T...T...-...... TT-.....A..------TT...T.TT...--.-..C.T.A..T...... G....C....ACT...AGA.CAT...... C...AT.AAGT.A..T...GT-..TTAA.....GG.CAA.C.....G 25 -...... C.....TT-.....A..------TT...C.TT...--.-..C.T.A..T...... G....C....AC..A.ATA..AG...... C...GT.AA.T.A..T...G.-..TTAA..T.....CAA...... G 26 -...... -C.....TT-.....A..------TT...C.TT...--.-..CCT.A..T...... G....C....AT..A.ATA..A..C....C...GT.AAGT.A..T...G.-..TTAA.T...... CAA...... G 27 -...... -

78 poly-A/poly-T rich segment (from residue 1-60). This is followed by a well conserved segment of 30 bases (residues 61-91) and another variable region incorporating poly-A/poly-

T mononucleotide repeats (residues 92-187). The 3’ end of the region (from bases 188-552) is highly conserved among quolls. The two segments incorporating the variable poly-A/poly-T repeats necessitated the insertion of a number of large indels for alignment purposes.

Species-specific length variations and/or base pair substitutions were found in the 5’ variable domain of the control region in each species (Figure 10). For instance, the total length of

Dasyurus maculatus sequences was 471 bp when unaligned, D. viverrinus sequences ranged from 488-489 bp, D. hallucatus sequences were between 480-484 bp, and the sole

D. albopunctatus sequence was 496 bp in length. Both D. geoffroii and D. spartacus sequences were within the range of that found in D. hallucatus (482 bp), but the many unique base pair substitutions readily identify the sequences to species.

Results of parsimony analyses are shown in a majority rule consensus tree (Figure 11) of 33 most parsimonious trees (524 steps; consistency index = 0.740; homoplasy index = 0.260; retention index = 0.829). There was moderate support for monophyly of Dasyurus as a sister taxon to Sarcophilus (74% of bootstrap replicates). As noted above, all species of quolls were readily identifiable from their unique control region sequences; this was upheld by the strongly supported species clades resolved in this tree. All sequences of D. geoffroii, D. hallucatus, and D. viverrinus with the exception of sequence 18 (TV.E11) for which there was limited sequence available, formed their

79 Figure 11. Majority rule consensus tree of 33 equally most parsimonious trees, based on 148 parsimony informative sites in the mtDNA control region. Values above nodes are bootstrap percentages out of 500 replicates (only values greater than 50% are shown), values below nodes are the percent of trees in which the branch was present in less than 100% of trees (all other nodes were present in all 33 trees). Sequences are numbered as in Table 3. D.a. = Dasyurus albopunctatus, D.g. = Dasyurus geoffroii, D.h. = Dasyurus hallucatus, D.m. = Dasyurus maculatus, D.v. =Dasyurus viverrinus, and D.s. = Dasyurus spartacus. 85 99 19 D. g. 56 20 21 D. s. 22 D. a. 13 98 14 53 15 D. v. 53 16 17 18 92 59 1 9 76 8 Mainland 53 7 D. m. 10 73 11 100 2 74 55 12 3 89 82 5 Tas D. m. 4 98 6 66 23 63 25 NT D. h. 80 27 59 26 100 79 29 28 24 Qld D. h. 30 Sarcophilus 31 Myrmecobius 32 Isoodon

80 own highly supported clades. (TV.E11 was subsequently removed from further analysis due to the large number of missing characters.) Similarly, all D. maculatus individuals form a clade which occurred in 100% of bootstrap replicates. However, within D. maculatus there was a strongly supported clade of Tasmanian tiger quolls separate from mainland individuals, and within D. hallucatus there was a strongly supported clade of Northern Territory individuals separate from Queensland individuals.

While there was strong support for each species clade and relationships among most quoll species were resolved, these relationships were not highly supported. However, a sister group relationship between D. geoffroii and D. spartacus was supported in 99% of bootstrap replicates, and there was some support (55%) suggesting a sister relation between the D. geoffroii + D. spartacus clade and D. albopunctatus.

The neighbor-joining tree obtained from Tamura distances is shown in Figure 12. The topology of this tree is essentially identical to that of the maximum parsimony tree except for minor rearrangements of the terminal taxa within the species clades. This analysis provides somewhat stronger bootstrap support (79%) for the sister group relation of D. albopunctatus to D. spartacus + D. geoffroii. The removal of the incomplete sequence for individual 18

(TV.E11) from the neighbor-joining analysis shows the highly supported clade of D. viverrinus. In addition, there is again strong support for separate clades of Tasmanian and mainland D. maculatus, as well as Northern Territory and Queensland D. hallucatus.

Finally, resolution of D. hallucatus as the sister species to all other Dasyurus is evident.

81 Figure 12. Neighbor-joining tree based on pairwise Tamura distances. Values at major nodes are bootstrap percentages out of 500 replicates (values at minor nodes not shown). Sequences are numbered as in Table 3. Abbreviations are as in Figure 11.

78 19 78 100 D. g. 20 21 D. s. 44 22 D. a. 14 98 13 17 D. v. 15 74 16 1 79 9 51 8 Mainland 45 11 81 D. m. 100 7 10 2 12 4 6 Tas D. m. 98 3 5 23 25 96 100 27 28 NT D. h. 26 29 24 Qld D. h. 30 Sarcophilus 31 Myrmecobius 32 Isoodon

Scale: is @ to the distance of 0.04

82 Tamura distances among conspecifics were low in comparison to those among species (Table

4). Values ranged from 0.002-0.042 among D. maculatus; 0.000-0.056 among D. viverrinus; 0.019 among D. geoffroii; and 0.004-0.046 among D. hallucatus. With the exception of distances between D. geoffroii and D. spartacus, values among species ranged from 0.064-0.180. Tamura distances between D. geoffroii and D. spartacus fell within the range of conspecific values (0.023-0.030).

Results of the Kishino-Hasegawa test are shown in Table 5. The phylogenetic tree proposed by Van Dyke (1987) was 12 steps longer the currently proposed topology, and was significantly worse by this test. Similarly, the topology proposed by Wroe and Mackness

(1998) was 13 steps longer than the currently proposed topology and was also significantly worse. Phylogenetic relationships based on morphological data is not supported by the mtDNA control region data.

A matrix of the total number of nucleotide transitions and transversions was calculated using the pairwise-deletion option in MEGA (Kumar et al., 1993) (Table 4). These values were used to generate the minimum spanning network (MSN) shown in Figure 13. This ‘tree’ shows the relationships of the haplotypes based upon the minimum number of substitutions.

As in both the parsimony tree and the neighbor-joining tree, the MSN illustrates the high degree of similarity among haplotypes within species and also the great degree of differentiation between species. The average number of within species substitutions among pairs of D. viverrinus sequences is 2.8 bases when sequence 18 (TV.E11) is excluded.

Within Northern Territory D. hallucatus, the average number of substitutions is 5.9 bases; among Tasmanian and mainland

83 1 2 3 15 16 1 0 4 5 14 1 43 33 40 42 4544 3643 35 42 34 41 44 40 43 42 9 12 41 32 3811 40 12 4613 41 37 42 37 43 33 39 45 39 40 41 3 8 8 7 4 57 6 6 10 11 12 13 9 7 6 5 11 10 1 9 9 9 6 1 6 5 1919 1818 1718 17 17 17 17 16 16 16 16 15 15 15 15 14 14 14 14 43 13 45 13 34 42 35 44 40 33 42 35 42 39 44 41 41 43 2 5 5 6 1 3 14 14 13 13 11 10 16 16 15 15 1 2 3 4 5 6 7 8 0.022 0.035 0.031 0.035 0.0310.033 0.013 0.0280.033 0.002 0.028 0.011 0.0060.020 0.011 0.0240.014 0.004 0.042 0.021 0.0420.002 0.042 0.040 0.024 0.0400.013 0.040 0.037 0.040 0.017 0.0380.011 0.040 0.035 0.035 0.021 0.015 0.035 0.0350.026 0.033 0.033 0.022 0.009 0.033 0.033 0.016 0.0310.096 0.031 0.011 0.031 0.101 0.031 0.0140.082 0.028 0.101 0.013 0.015 0.085 0.028 0.106 0.0110.087 0.088 0.024 0.099 0.013 0.092 0.090 0.0280.092 0.104 0.006 0.092 0.085 0.029 0.109 0.097 0.097 0.0910.093 0.017 0.102 0.097 0.089 0.096 0.096 0.020 0.099 0.1020.098 0.094 0.097 0.096 0.107 0.094 0.100 0.102 0.085 0.099 0.104 0.099 0.096 0.0890.104 0.094 0.093 0.101 0.104 0.089 0.095 0.099 0.091 0.099 0.0990.106 0.097 0.092 0.104 0.088 0.108 0.094 0.101 0.094 0.098 0.115 0.000 0.1090.106 0.102 0.106 0.092 0.115 0.107 0.096 0.101 0.099 0.117 0.113 0.0090.109 0.112 0.105 0.106 0.097 0.109 0.005 0.102 0.102 0.104 0.102 0.116 0.0110.133 0.114 0.105 0.106 0.101 0.099 0.108 0.007 0.109 0.130 0.105 0.107 0.109 0.007 0.002 0.1140.144 0.108 0.141 0.105 0.094 0.104 0.007 0.103 0.141 0.109 0.146 0.041 0.102 0.109 0.0070.137 0.100 0.147 0.101 0.138 0.041 0.094 0.112 0.010 0.135 0.108 0.153 0.104 0.050 0.1440.141 0.107 0.087 0.146 0.101 0.144 0.101 0.127 0.053 0.067 0.138 0.112 0.151 0.1030.131 0.150 0.139 0.092 0.084 0.149 0.110 0.143 0.095 0.138 0.128 0.072 0.135 0.087 0.155 0.1070.135 0.148 0.084 0.151 0.139 0.089 0.130 0.147 0.104 0.132 0.132 0.064 0.147 0.145 0.091 0.1300.141 0.152 0.085 0.142 0.143 0.081 0.141 0.136 0.124 0.135 0.078 0.139 0.140 0.149 0.079 0.141 0.1420.222 0.130 0.145 0.082 0.150 0.135 0.140 0.130 0.125 0.121 0.219 0.144 0.087 0.156 0.135 0.1460.302 0.139 0.147 0.130 0.219 0.138 0.147 0.129 0.129 0.292 0.139 0.133 0.133 0.231 0.138 0.1530.403 0.139 0.137 0.312 0.144 0.128 0.222 0.133 0.136 0.130 0.388 0.138 0.316 0.147 0.128 0.228 0.145 0.138 0.138 0.404 0.132 0.312 0.122 0.222 0.130 0.144 0.140 0.403 0.132 0.319 0.128 0.204 0.138 0.138 0.399 0.127 0.305 0.121 0.225 0.141 0.138 0.407 0.137 0.291 0.128 0.222 0.133 0.395 0.130 0.305 0.131 0.222 0.144 0.394 0.138 0.302 0.225 0.136 0.407 0.141 0.302 0.169 0.144 0.391 0.295 0.151 0.147 0.391 0.318 0.174 0.388 0.291 0.177 0.406 0.307 0.382 0.307 0.406 0.403 Complete sequence unavailable. Boxed areas denote distances within major monophyletic groups. Table 4. Pairwise Tamura distance (below diagonal) and total number of observed substitutions (pairwise deletion method; above diagonal). Sequences are numbered as in Table 3. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 1

84 166 133 158 80 98 144 3 82 97 144 4 7 78 96 139 7 3 4 81 96 145 10 5 7 10 82 97 145 20 21 16 18 19 81 101 144 18 3 9 2 6 9 80 95 141 74 75 75 74 72 75 74 75 104 146 31 56 59 57 56 54 57 58 79 101 146 14 34 59 63 60 60 58 60 62 82 106 144 9 11 33 57 60 58 58 56 58 60 82 105 140 19 20 21 22 23 24 25 26 27 28 2925 30 27 31 25 32 28 41 44 42 42 41 42 42 45 77 101 1 20 32 34 31 34 49 55 52 52 49 52 54 58 102 124 18 1 333 15 154 37 18 26 19 39 37 28 38 36 39 25 40 36 36 30 35 53 36 45 38 59 54 51 55 56 59 48 60 56 57 48 58 52 57 45 58 56 53 48 54 58 57 50 58 70 59 57 60 119 72 101 142 73 122 116 144 116 143 3536 3436 3537 44 3135 42 3337 45 44 3240 43 48 34 4540 45 45 39 4336 49 47 46 38 4539 46 43 44 35 4938 48 54 42 43 36 4637 46 53 45 46 36 48 58 43 57 40 44 36 44 57 46 59 43 46 56 40 59 43 56 40 47 55 46 61 44 58 42 57 59 43 58 43 52 47 56 61 44 60 53 54 46 60 58 41 56 52 55 45 62 60 58 55 56 53 44 59 54 59 54 58 53 61 57 55 57 58 55 51 55 56 57 57 60 57 87 56 60 55 53 57 51 86 58 62 55 113 59 54 86 56 59 53 140 110 53 54 90 56 61 116 136 54 52 87 54 55 140 117 56 52 89 56 140 116 54 50 87 58 118 139 54 77 56 141 114 52 88 56 138 105 87 54 131 114 87 113 141 88 137 113 137 111 136 17 0.056 0.086 0.071 0.091 0.077 0.019 0.082 0.0710.093 0.023 0.079 0.030 0.0750.137 0.077 0.1200.156 0.070 0.138 0.129 0.1430.146 0.147 0.134 0.123 0.154 0.176 0.146 0.140 0.143 0.123 0.145 0.1800.137 0.141 0.136 0.039 0.120 0.145 0.1780.146 0.136 0.134 0.006 0.123 0.140 0.1760.152 0.043 0.140 0.130 0.019 0.123 0.145 0.172 0.046 0.1460.160 0.137 0.004 0.021 0.151 0.133 0.178 0.0350.308 0.140 0.206 0.013 0.011 0.244 0.177 0.205 0.0390.405 0.015 0.278 0.019 0.195 0.015 0.349 0.279 0.041 0.186 0.006 0.399 0.263 0.021 0.205 0.008 0.411 0.277 0.0080.209 0.417 0.253 0.015 0.210 0.416 0.273 0.006 0.208 0.403 0.258 0.200 0.417 0.256 0.210 0.417 0.257 0.205 0.419 0.259 0.398 0.262 0.413 0.351 0.416 0.443 0.481 Complete sequence unavailable. Boxed areas denote distances within major monophyletic groups. Table 4. Continued 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 1

85 Yes Yes worse? Best tree Significantly - Standard deviation 0.00 24.61 10.99 30.71 13.96 in scores Difference score -1604.37 -1628.98 -1635.08 Maximum likelihood 1 1 1 examined No. of trees 208 220 221 in tree No. of steps Proposed Topology Firestone, 1999 (Figure 11) Van Dyke, 1987 (Figure 7) Wroe and Mackness, 1998 (Figure 9) Table 5. Kishino-Hasegawa (1989) test results

86 Figure 13. Minimum spanning network based on the number of base pair substitutions between pairs of sequences. Only pairs of sequences with greater than 10 base pair substitutions are indicated by numbers on branches. Sequences are numbered as in Table 3. 31 Myrmecobius

24 Queensland 95 16 D. hallucatus 27 26 Sarcophilus 23 28 Northern Territory 30 25 57 45 29 D. spartacus 21 19 11 20 15 16 14 25 30 D. geoffroii 13 17 D. viverrinus 32 22 D. albopunctatus D. maculatus 9 1

12 7 11 13 10 4 mainland 2 8 13 5 6 3 Tasmania

87 D. maculatus haplotypes there is an average of 3.7 and 7.7 base substitutions, respectively.

The number of substitutions between species of quolls is large in comparison to that within species. For example, the minimum and average number of substitutions between

D. viverrinus and D. hallucatus are 45 and 53.9, respectively. The comparable values between D. viverrinus and D. maculatus are 32 and 39.5, respectively, and between D. viverrinus and D. albopunctatus are 30 and 34.8 respectively (Table 4,

Figure 13). A notable exception is found between D. spartacus and D. geoffroii, where sequences differ by an minimum of 11 (average 12.5) substitutions. This is low in comparison to the differentiation observed between mainland and Tasmanian forms of

D. maculatus (minimum 13, average 15.6 substitutions) or between Queensland and

Northern Territory forms of D. hallucatus (minimum 16, average 18.7 substitutions).

Discussion

Features of the Control Region in Dasyurus

The extremely high AT content observed in the left domain of the control region of quolls is in keeping with that found in other mammalian species (Wolstenholme, 1992), although the length of this region in quolls was among the shortest found within marsupials (L. Fumagalli, pers. comm.). The 5’ poly-A/poly-T rich segment is variable in the number of adenine and thymine residues both within species and between species.

Mutations at this site may arise possibly by slippage, as occurs in other mitochondrial domains and microsatellite loci (Schlötterer and Tautz, 1992; Fumagalli et al., 1996).

88 The overall high level of variability in the 5’ segment of the control region accords with that found in many mammalian orders (Lopez et al., 1997).

Each species of quoll possessed unique identifiers, in sequence length and/or in particular base pair substitutions, that allowed ready identification to a particular species.

These synapomorphies in the control region sequences of each species may provide a simple, reliable, and relatively inexpensive diagnostic test in forensic applications.

Phylogenetic Relationships Within Dasyurus

The phylogenetic reconstruction methods used here (minimum evolution and maximum parsimony) are based on different assumptions of evolution. The high degree of concordance between the two trees at all major nodes (Figures 11 and 12) is considered to reflect strong support for the phylogenetic branching pattern.

The control region sequence data presented here lend additional support to the phylogenetic reconstruction derived from other molecular studies. Specifically, our findings concur with Krajewski et al. (1997) in the placement of: (1) Dasyurus geoffroii and D. spartacus as sister species, (2) D. albopunctatus as sister to the D. geoffroii + D. spartacus clade, (3) D. hallucatus as a sister to all other quolls, and (4)

Sarcophilus as a sister to Dasyurus. However, it should be noted that some taxa were sampled more extensively and that this may bias the results. In addition, the use of distant outgroups such as Myrmecobius and Isoodon may have the effect of incorrectly rooting the tree and providing a spurious ingroup, i.e. showing that Dasyurus is

89 monophyletic with respect to Sarcophilus when this might not be the case. Despite the use of distant outgroups, the data presented here concur with other reconstructions in placing Sarcophilus as a sister taxon to Dasyurus (Archer, 1976b; Krajewski et al.,

1997).

The relationships of D. viverrinus and D. maculatus to the other species of Dasyurus were resolved but not highly supported by bootstrap replicates. The lack of high support may be due either to the relatively recent species radiation among quolls [~ 10 million years ago, based on the immunological data of Baverstock et al. (1990)], or the rapidity of cladogenesis.

Corroboration between the reconstruction of Krajewski et al. (1997) and that presented here might not be surprising given that the loci examined are linked on the same mitochondrial genome. However, Krajewski et al. (1997) also included an independent nuclear gene, protamine P1, in their overall reconstruction. In addition, these two molecular studies are the first to concur in the overall topology. No other previous reconstructions using similar morphological techniques (e.g. dental or basicranial characters) have resulted in similar topologies.

Results presented here call into question the degree of differentiation and the separate specific status conferred upon D. geoffroii and D. spartacus. Examination of sequence data indicates that differences between these two species are extremely low and within the range of intraspecific comparisons (Table 4). Krajewski et al. (1997) also reported very low sequence divergence values for D. geoffroii and D. spartacus. Uncorrected

90 ‘p-values’ averaged 2.6% (range 2.3-2.9%) between D. geoffroii and D. spartacus for the control region (these values were 7.3%, 1.5%, and 0.0% for cytochrome b, 12S rRNA, and protamine P1, respectively; C. Krajewski, pers. comm.). This low level of differentiation is of particular interest since D. spartacus was only recently recognized as a species based on morphological distinctions and was previously thought to be an extralimital population of D. geoffroii (Waithman, 1979; Van Dyck, 1987).

Supplementary sequence data from additional individuals and additional loci may help clarify the distinctions between these two species.

The extremely close molecular relationship between D. geoffroii and D. spartacus is puzzling in terms of both the current biogeographic distributions and habitat adaptations of these species. Biogeographically, D. hallucatus would appear to be the most likely disperser to New Guinea since this species occurs widely throughout the northern third of the Australian continent, including the Cape York peninsula (Strahan, 1998), which is less than 150 kilometers from New Guinea. Additionally, Flannery (1995) notes that the absence of D. spartacus from the tip of Cape York is enigmatic. D. geoffroii, however, is currently limited to the southwestern corner of the Australian continent, approximately 4000 kilometers away from the New Guinean mainland. In terms of habitat preference, D. spartacus occurs in low mixed savannah (Flannery, 1995), similar to the habitat of D. hallucatus in the Cape York peninsula, whereas D. geoffroii is an arid-adapted species. Yet D. hallucatus is clearly distinct genetically from D. spartacus. The high degree of genetic similarity between D. geoffroii and D. spartacus may be partially explained by examining the recent historic distribution of D.

91 geoffroii. Although the fossil record is lacking, D. geoffroii was much more widespread historically than it is today, occurring in both Queensland and the Northern

Territory (Serena et al., 1991). In addition, there are numerous instances of closely related mesic-xeric species pairs among the dasyurids.

Phylogenetic Relationships Within Species

Results of maximum parsimony and minimum evolution algorithms showed distinct phylogenetic lineages within species where geographically separate regions were sampled. MtDNA genes have a lower effective size than nuclear genes due to their uniparental inheritance (Birky et al., 1989). The resulting rapid drift, together with lack of recombination and possibly selection, leads to relatively rapid divergence of mtDNA lineages compared to that of nuclear DNA.

Within D. maculatus, Tasmanian animals constitute a phylogeographically distinct lineage separate from those on the mainland. Similarly, within D. hallucatus,

Queensland and Northern Territory individuals comprise distinct lineages. Distinctions between mainland and Tasmanian D. maculatus are evident in nuclear as well as mitochondrial loci and in certain morphological characters (Firestone et al., 1999;

Jones, 1997), warranting review of the taxonomic status within this species, and elevation to subspecific status for Tasmanian D. maculatus.

In the past, four subspecies of D. hallucatus were recognized on the basis of morphological differences and geographical location (Gould, 1842; Thomas, 1909;

92 Thomas, 1926): D. h. hallucatus (Northern Territory), D. h. nesaeus (Groote

Eylandt), D. h. exilis (Western Australia), and D. h. predator (Cape York

Peninsula, Queensland). However these trinomials are no longer current (e.g. Strahan,

1998; Maxwell et al., 1996). Furthermore, the lack of taxonomic clarity has proven to be a major stumbling block in the conservation of many species (e.g. Daugherty et al.,

1990; Zink and Kale, 1995). This is the first study to examine geographically disjunct populations of D. hallucatus using molecular techniques. Preliminary results suggest that there are strongly divergent lineages among D. hallucatus from the Northern

Territory and Queensland; sequences between the geographic groups are at least as divergent as those between D. geoffroii and D. spartacus. Further studies are focussing on the remaining geographically disjunct populations; a resurrected subspecific classification may be necessary for D. hallucatus.

References

Anderson S, Bankier AT, Barrell BG, De Bruijn MHL, Coulson AR, Drouin J, Eperon

IC, Nierlich DP, Roe BA, Sanger F, Schreier PH, Smith AJH, Staden R, Young IG

(1981) Sequence and organization of the human mitochondrial genome. Nature, 290,

457-65.

Aplin KP, Archer M (1987) Recent advances in marsupial systematics with a new syncretic classification. In: Possums and Opossums: Studies in Evolution. (ed. Archer

M). pp. xv-lxxii. Surrey Beatty and Sons and the Royal Zoological Society of New

South Wales, Sydney.

93 Archer M (1976a) The basicranial region of marsupicarnivores (Marsupialia), interrelationships of carnivorous marsupials, and affinities of the insectivorous marsupial peramelids. Zoological Journal of the Linnean Society, 59, 217-322.

Archer M (1976b) The dasyurid dentition and its relationships to that of didelphids, thylacinids, borhyaenids (Marsupicarnivora) and peramelids (Peramelida:Marsupialia).

Australian Journal of Zoology, Suppl. Ser. 39, 1-34.

Archer M (1982) Review of the dasyurid (Marsupialia) fossil record, integration of data bearing on phylogenetic interpretation, and suprageneric classification. In:

Carnivorous Marsupials (ed. Archer M) pp. 397-443. Royal Zoological Society of

New South Wales, Sydney.

Archer M (1984) The Australian marsupial radiation. In: Vertebrate Zoogeography and Evolution in Australasia, Animals in Space and Time (eds. Archer M, Clayton

G). pp. 633-808. Hesperian Press, Carlisle.

Baverstock PR, Archer M, Adams M, Richardson BJ (1982) Genetic relationships among 32 species of Australian dasyurid marsupials. In: Carnivorous Marsupials (ed.

Archer M) pp. 641-650. Royal Zoological Society, New South Wales, Sydney.

Baverstock PR, Krieg M, Birrell J (1990) Evolutionary relationships of Australian marsupials as assessed by albumin immunology. Australian Journal of Zoology, 37,

273-287.

94 Birky CW, Fuerst P, Maruyama T (1989) Organelle gene diversity under migration, mutation, and drift: equilibrium expectations, approach to equilibrium, effects of heteroplasmic cells, and comparison to nuclear genes. Genetics, 121, 613-627.

Campbell NJH, Harriss FC, Elphinstone MS, Baverstock PR (1995) Outgroup heteroduplex analysis using temperature gradient gel electrophoresis: high resolution, large scale, screening of DNA variation in the mitochondrial control region. Molecular

Ecology, 4, 407-418.

Daugherty CH, Cree A, Hay JM, Thompson MB (1990) Neglected taxonomy and continuing extinctions of tuatara (Sphenodon). Nature, 347, 177-179.

Elphinstone MS, Baverstock PR (1997) Detecting mitochondrial genotypes by temperature gradient gel electrophoresis and heteroduplex analysis. BioTechniques, 23,

982-986.

Excoffier L, Smouse PE (1994) MINSPNET. Available from http://anthropologie. unige.ch/LGB/software/win/min-span-net/

Firestone K, Elphinstone M, Sherwin W, Houlden B (1999) Phylogeographical population structure of tiger quolls Dasyurus maculatus (Dasyuridae:Marsupialia), an endangered carnivorous marsupial. Molecular Ecology, 8, 1613-1626.

95 Flannery T (1995) Mammals of New Guinea, Reed Books, Sydney.

Fumagalli L, Pope LC, Taberlet P, Moritz C (1997) Versatile primers for the amplification of the mitochondrial DNA control region in marsupials. Molecular

Ecology, 6, 1199-1201.

Fumagalli L, Taberlet P, Favre L, Hausser J (1996) Origin and evolution of homologous repeated sequences in the mitochondrial DNA control region of shrews. Molecular

Biology and Evolution, 13, 31-46.

Gould J (1842) Characters of a new species of Perameles, and a new species of

Dasyurus. Proceedings of the Zoological Society of London, 1842, 41-42.

Gyllensten U, Wharton D, Josefsson A, Wilson AC (1991) Paternal inheritance of mitochondrial DNA in mice. Nature, 352, 255-257.

Jones M (1997) Character displacement in Australian dasyurid carnivores: size relationships and prey size patterns. Ecology, 78, 2569-2587.

Kirsch JAW (1977) The comparative serology of Marsupialia, and a classification of marsupials. Australian Journal of Zoology, Suppl. Ser. 52, 1-152.

Kirsch JAW, Archer M (1982) Polythetic cladistics, or, when parsimony's not enough: the relationships of carnivorous marsupials. In: Carnivorous Marsupials,

96 (ed. Archer M) pp. 595-619. Royal Zoological Society of New South Wales, Sydney.

Kirsch JA, Dickerman AW, Reig OA, Springer MS (1991) DNA hybridization evidence for the Australian affinity of the American marsupial Dromiciops australis.

Proceedings of the National Academy of Sciences USA, 88, 10465-10469.

Kirsch JAW, Krajewski C, Springer MS, Archer M (1990a) DNA-DNA hybridisation studies of carnivorous marsupials. II. Relationships among dasyurids

(Marsupialia:Dasyuridae). Australian Journal of Zoology, 38, 673-696.

Kirsch JAW, Springer MS, Krajewski C, Archer M, Aplin K, Dickerman AW

(1990b) DNA/DNA hybridization studies of the carnivorous marsupials. I: The intergeneric relationships of bandicoots (Marsupialia:Perameloidea). Journal of

Molecular Evolution, 30, 434-448.

Kishino H, Hasegawa M (1989) Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in

Hominoidea. Journal of Molecular Evolution, 29, 170-179.

Krajewski C, Painter J, Driskell AC, Buckley L, Westerman M (1993) Molecular systematics of New Guinean dasyurids (Marsupialia:Dasyuridae). Science in New

Guinea, 19, 157-166.

97 Krajewski C, Painter J, Buckley L, Westerman M (1994) Phylogenetic structure of the marsupial family Dasyuridae based on cytochrome b DNA sequences. Journal of

Mammalian Evolution, 2, 25-35.

Krajewski C, Young J, Buckley L, Woolley PA, Westerman M (1997) Reconstructing the evolutionary radiation of dasyurine marsupials with cytochrome b, 12S rRNA, and protamine P1 gene trees. Journal of Mammalian Evolution, 4, 217-236.

Kumar S, Tamura K, Nei M (1993). MEGA: Molecular Evolutionary Genetics

Analysis. The Pennsylvania State University, University Park, PA 16802.

Lopez JV, Culver M, Stephens JC, Johnson WE, O'Brien, SJ (1997) Rates of nuclear and cytoplasmic mitochondrial DNA sequence divergence in mammals. Molecular

Biology and Evolution, 14, 277-286.

Madison WP, Madison DR (1992) Maclade. Version 3.01. Sinauer Associates,

Sunderland, Massachusetts.

Mahoney JA, Ride WDL (1984) The identity of Captain Cook's quoll Mustela quoll

Zimmermann, 1783 (Marsupialia: Dasyuridae). Australian Mammalogy, 7, 57-62.

98 Marshall LG, Case JA, Woodburne MO (1990) Phylogenetic relationships of the families of marsupials. In: Current Mammalogy, (ed. Genoways HH) pp. 433-505.

Plenum Press, New York.

Maxwell S, Burbidge AA, Morris K, Eds. (1996). The 1996 Action Plan for

Australian Marsupials and Monotremes. Wildlife Australia, Canberra.

Pocock RI (1926) The external characters of Thylacinus, Sarcophilus, and some related marsupials. Proceedings of the Zoological Society of London, 1926, 1037-

1084.

Sambrook E, Fritsch F, Maniatis T (1989) Molecular Cloning: A Laboratory

Manual, 2nd edn. Cold Spring Harbor Laboratory Press, New York.

Schlötterer C, Tautz D (1992) Slippage synthesis of simple sequence DNA. Nucleic

Acids Research, 20, 211-215.

Serena M, Soderquist TR, Morris K (1991) Western Austalian Wildlife Management

Program No. 7. The Chuditch (Dasyurus geoffroii), Department of Conservation and Land Management, Como WA

Springer MS, Westerman M, Kirsch JAW (1994) Relationships among orders and families of marsupials based on 12S ribosomal DNA sequences and the timing of the marsupial radiation. Journal of Mammalian Evolution, 2, 85-115.

99 Strahan R, Ed. (1998) The Mammals of Australia. New Holland Publishers, Sydney.

Strimmer K, von Haeseler A (1996) Quartet puzzling: a quartet maximum likelihood method for reconstructing tree topologies. Molecular Biology and Evolution, 13, 964-

969.

Swofford DL (1999) PAUP* Phylogenetic Analysis Using Parsimony (*and Other

Methods) Version 4. Sinauer Associates, Sunderland, Massachusetts.

Tamura K (1992) Estimation of the number of nucleotide substitutions when there are strong transition-transversion and G+C-content biases. Molecular Biology and

Evolution, 9, 678-687.

Thomas O (1909) Some mammals from N.E. Kimberley, Northern Australia. The

Annals and Magazine of Natural History; Zoology, Botany, and Geology, 8, 149-

152.

Thomas O (1926) The local races of Dasyurus hallucatus. The Annals and Magazine of Natural History; Zoology, Botany, and Geology, 9, 543-544.

Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions- specific gap penalties and weight matrix choice. Nucleic Acids Research, 22, 4673-

4680.

100 Van Dyck S (1987) The bronze quoll, Dasyurus spartacus (Marsupialia:Dasyuridae), a new species from the savannahs of Papua New Guinea. Australian Mammalogy, 11,

145-156.

Waithman J (1979) A report on a collection of mammals from Southwest Papua, 1972-

1973. Australian Zoologist, 20, 313-326.

Wiens JJ, Servedio MR (1997) Accuracy of phylogenetic analysis including and excluding polymorphic characters. Systematic Biology, 46, 332-345.

Wolstenholme DR (1992) Animal mitochondrial DNA: structure and evolution.

International Review of Cytology, 141, 173-216.

Wroe S, Mackness BS (1998) Revision of the Pliocene dasyurid, Dasyurus dunmalli,

(Dasyuridae; Marsupialia). Memoirs of the Queensland Museum, 42, 605-612.

Zink RM, Kale HW (1995) Conservation genetics of the extinct dusky seaside sparrow

Ammodramus maritimus nigrescens. Biological Conservation, 74, 69-71.

Acknowledgments

I greatly appreciate the assistance of Luca Fumagalli who provided the control region primers and the sequences of the bandicoot and numbat prior to publication, and the many colleagues and institutions who provided samples used in this study. Martin

101 Elphinstone and Dean Jerry instructed me in the use of temperature gradient gels;

Margaret Heslewood and Chris Quinn provided analytical assistance with PAUP; Joe

Zuccarello kindly provided help with the Kishino-Hasegawa test; and Steve Wroe initially gave me the idea of using these data for phylogenetic analyses. I also wish to thank Warwick Greville, Carey Krajewski, and Mike Westerman for helpful discussions and Chris Quinn, Chris Dickman, Jack Giles, and Bill Sherwin for providing useful comments on an earlier draft of the manuscript. Carey Krajewski, Mark Springer, and

Patrick Luckett also provided valuable comments. This work was supported by an

Australian Postgraduate Research Award, the W.V. Scott Foundation, Barrington

Guest House, and the Zoological Parks Board of New South Wales.

102 Chapter 4. Variability and Differentiation: Microsatellites in the

Genus Dasyurus and Conservation Implications for the Large

Australian Carnivorous Marsupials

All four species of Australian quolls (Dasyurus species) have declined since European settlement in terms of both range and population numbers. Six highly polymorphic simple

sequence repeats (CAn microsatellites) were used to estimate the genetic variability and population differentiation within and among twenty populations (including museum specimens from six populations), as a preliminary means of assessing population conservation status and relative levels of variability within members of the genus. Overall mean expected

heterozygosity (HE) and corrected allelic diversity (A’) were highest among western quolls.

Northern quolls, eastern quolls, and tiger quolls were not significantly different from each other in either measure. There were also significant differences in diversity among populations within species. Genetic differentiation was estimated by a number of methods and showed that the microsatellites used here were useful for defining differences both among species and populations. Allele frequency data were summarised by two-dimensional MDS, which was able to partition populations into distinct species clusters. Similarly, the assignment test was able to assign most individuals to both the correct species and population levels. Results of

MDS and the assignment test may prove useful in forensic applications. Genetic distance and subdivision between pairs of populations were assessed by two means based on different

mutation models for microsatellites: infinite alleles model (Nei’s D, FST) and stepwise mutation

2 model (Goldstein’s dm , RST). Pairwise measures of population subdivision indicate that most populations should be conserved as separate management units. I discuss results of these

103 analyses in terms of applications to conservation for each of the four Australian species of quoll and provide a genetic basis for future population monitoring in these species.

Introduction

Genetic variability or diversity has long been recognized as a key component of population and conservation genetics. The loss of genetic variation (either allelic diversity or heterozygosity), due to drift, inbreeding, or other factors can reduce both individual fitness and the ability of populations to adapt to altered environmental conditions (Lacy 1997). Inbreeding may cause decreased levels of heterozygosity in individuals. Inbreeding depression, as a result of close consanguineous matings, may decrease individual fitness via a number of routes (Gall 1987;

Ralls et al. 1988; Frankham 1995; Newman & Pilson 1997). The long term viability of populations also may be threatened by the loss of genetic variation due to drift. Loss of variability may affect small populations by decreasing the ability to adapt to changing environments and by increasing the probability of extinction due to stochastic effects (Lacy

1997). Alternatively, relatively high levels of genetic diversity may increase qualities associated with fitness (Allendorf & Leary 1986).

Differentiation between populations is another area of concern to population and conservation geneticists, as this gives an indication of the evolutionary divergence between populations.

Much debate has been focussed on determining what units to conserve, particularly given limited resources (Vane-Wright et al. 1991; Crozier 1992; Rojas 1992; Vogler & DeSalle

1994; Waples 1998). Moritz (1994) and Moritz et al.

104 (1995) suggested that populations that are genetically divergent at both nuclear and mitochondrial loci should be conserved as separate units, i.e. evolutionarily significant units

(ESUs) or management units (MUs). Several recent studies have applied these concepts to conservation of endangered taxa (e.g. Pope et al. 1996; Zhu et al. 1998; Firestone et al.

1999).

Analysis of microsatellites has become an important tool in population studies and is useful for estimating both variability and differentiation (e.g. Taylor et al. 1994; FitzSimmons et al.

1995; García-Moreno et al. 1996; Houlden et al. 1996; Brunner et al. 1998).

Microsatellites are single-locus, biparentally inherited, and highly variable markers occurring throughout the genome (Tautz 1989). They contain short-length (usually di-, tri-, or tetra- nucleotide) repeat units which vary in the number of repeats between individuals (Tautz &

Renz 1984; Tautz 1989). In addition, microsatellites have proven to be highly informative where other markers have yielded little information when applied to the same species (e.g.

Paetkau & Strobeck 1994; Estoup et al. 1998; Goodman 1998).

The six species of quolls (Dasyurus spp.) are among the largest of the remaining carnivorous marsupials in Australia and Papua New Guinea. The four species of quolls found in Australia range in size from seven kilograms (some male tiger quolls) to less than 400g (female northern quolls) (Strahan 1998). Like their placental counterparts, these large marsupial carnivores have faced declines in numbers and distribution throughout their ranges (Figure 14). The reasons for these declines are poorly understood, and are likely to be due to a number of interacting factors. Factors such as

105 Figure 14. Current (black) and past (grey + black) distribution of the Australian quolls including sample sites. A) Dasyurus maculatus, B) D. viverrinus, C) D. hallucatus, D) D. geoffroii. A, B, and D redrawn from (Strahan 1998); C redrawn from (Braithwaite & Griffiths 1994). See Table 6 for key to population names.

A B

MW

GI

BT CO BG CH NE BS WY SB ST GL TT VB TE

C D

AT KA AR

BF

PE

106 habitat loss, the introduction of feral predators and poisonous prey resources, altered fire regimes, disease susceptibility, and continued persecution by humans may all have played different roles in the decline of each species (Maxwell et al. 1996).

Each quoll species is considered to be threatened to some degree and each species has had a different history of decline. The western quoll (Dasyurus geoffroii) exhibited a population decline on a continental scale and survives only in one area of southwestern Western Australia

(WA). Due to recent successful management efforts, this species has been downgraded from

‘endangered’ to ‘vulnerable’ by the World Conservation Union (IUCN) (Serena et al. 1991;

Orell & Morris 1994; Maxwell et al. 1996). Eastern quoll (D. viverrinus) numbers were decimated around the turn of this century; they are currently presumed extinct on the mainland.

This species persists in Tasmania, however, where population numbers are stable. Eastern quolls are currently designated as ‘lower risk-near threatened’ (IUCN listing; Maxwell et al.

1996). The decline of northern quolls (D. hallucatus) is a relatively recent phenomenon.

Once widely distributed throughout the northern third of the continent, this species is now restricted to six main population centres (Braithwaite & Griffiths 1994) and is also currently designated as ‘lower risk-near threatened’ by the IUCN. The southern mainland subspecies of tiger quolls (D. maculatus maculatus) declined early this century, however unlike eastern quolls, tiger quolls persisted on the mainland where populations are scattered and numbers are low (Mansergh 1984). Tiger quolls also occur in Tasmania, where populations are thought to be naturally limited by competition from both Tasmanian devils (Sarcophilus harrisii) and eastern quolls (Jones 1995). Southern mainland tiger quolls are presently restricted to less than 50% of their range and are

107 listed as ‘vulnerable’ by the IUCN. The northern mainland subspecies of tiger quoll (D. m. gracilis) is centred in a few small localities in north Queensland and is currently considered

‘endangered’ by the IUCN (Maxwell et al. 1996). Despite these widespread and precipitous declines, little attention has been given to genetic implications for conservation management of these taxa.

The applications of genetic management to quoll conservation are multifaceted. Captive management and breeding of western quolls is part of this species recovery plan and has been successful for a number of years (Orell & Morris 1994), yet determination of the level of diversity of the captive population has not been attempted. This is important in light of the reintroduction program taking place as part of this species recovery plan. In addition, there are plans to reintroduce eastern quolls into areas of their former range on the mainland; it is important to evaluate the genetic variation and differentiation between different stocks before reintroducing animals to areas where remnant populations might persist. Without sound knowledge of the genetic diversity within and between the remaining Tasmanian populations of eastern quolls, it is difficult to determine which populations are valuable sources for reintroductions. Analysis of the mitochondrial DNA control region and microsatellites has proven to be very useful in elucidating conservation units among tiger quolls (Firestone et al.

1999) and may be so for northern quolls.

I undertook this study to determine the relative levels of variability and differentiation present both among species and among populations within species, to provide baseline information regarding variability within populations of each species for future

108 population monitoring, and to assist wildlife agencies and managers in making sound conservation decisions regarding these species. In particular, I test the following null hypotheses: 1) genetic diversity is the same among all populations within each species, 2) each species of quoll has the same genetic diversity, 3) there is no genetic differentiation between populations within each species, and 4) there is no genetic differentiation between species of

quolls. I examined six highly polymorphic (CA)n microsatellite markers and a total of 347 individuals, representing 20 populations and four species, as a means of assessing the genetic variability and differentiation in quolls and as a basis for future population monitoring and conservation breeding programs.

Materials and Methods

Study Populations, DNA Samples, and Population Screening

Tissue samples were collected from all Australian species of quolls representing 20 different populations (Figure 14, Table 6). Six of these sample populations (PE, AR, NE, ST, GL, and WY) were from dried skins held in museum collections; two populations (BF, GI) were from captive stock bred in captivity for a number of years; and four groups (TE, TT, NE, and

BT) were from samples collected opportunistically and represent sites encompassing broader geographic areas. Samples were either fresh tissues (skin, blood, liver, or muscle) from live- trapped or road-killed individuals, or dried preserved skins from museum specimens. DNA from fresh tissues was extracted according to standard protocols by phenol-chloroform extraction followed by ethanol precipitation (Sambrook et al. 1989). DNA from museum specimens was extracted by a modified guanidine thiocyanate method (Boom et al. 1990;

Hoss & Pääbo 1993). The

109 § 2.5 (0.2) 2.7 (0.4) 4.3 (0.5) 4.7 (0.6) 4.7 (0.5) 4.2 (0.9) 2.5 (0.3) 2.5 (0.2) 4.3 (1.1) 3.8 (0.6) 7.0 (1.4) 4.8 (0.8) 3.5 (0.8) 3.7 (0.6) 9.0 (1.7) 3.6 (0.7) 4.0 (1.0) 8.8 (0.8) 9.2 (0.7) 2.8 (0.5)* 5 9 6 3 9 6 NA 12 12 16 11 32 11 25 13 58 21 14 26 23 35 wild wild wild wild wild wild wild wild wild wild wild wild captive captive museum museum museum museum museum museum STATUS 07’ 32' 03' ° 15’ ° 44’ ° 59’ ° 08’ ° 09’ ° ° 57’ ° 00’ ° 48’ ° 58’ ° 44’ ° 35’ ° 12’ ° 14’ 33’ 36 ° 02’ 16 ° 04’ 29 ° 48’ 31 ° 31’ 32 ° 42’ 32 ° ° 22’ 36 ° 44’ 41 ° 01’ 37 ° 01’ 40 ° 53' 41 ° 12’ 12 ° 09’ 13 ° 35' 17 ° 10’ 32 ° 13’ 33 EAST SOUTH MAP REFERENCE 145 152 151 151 151 wide area 149 148 145 wide area wide area 145 148 145 wide area 132 142 145 116 116 GI BT TT ST TE AT BF BS SB NE GL VB KA AR PE CO CH BG WY MW POP LOCATION 1. Mount Windsor, Qld 2. Glenn Innis, NSW 3. Copeland, NSW 4. Chichester State Forest, NSW 5. Barrington Guest House, NSW 6. Barrington Tops area, NSW 7. Badja State Forest, NSW 8. Suggan Buggan, Vic 9. Wynyard, Tas 10. Central Tasmania 11. New South Wales 12. Studley Park, Vic 13. Gladstone, Tas 14. Vale of Belvoir, Tas 15. Central Tasmania 16. Kakadu National Park, NT 17. Archer River, Qld 18. Atherton Tableland, Qld 19. Perth area, WA 20. Batalling State Forest, WA Five loci analysed. All other populations typed at all six loci. § SPECIES Tiger quoll Dasyurus maculatus Eastern quoll Dasyurus viverrinus Northern quoll Dasyurus hallucatus Western quoll Dasyurus geoffroii Table 6. Species and populations sampled. N = number of individuals sampled per population; A mean uncorrected allelic diversity, standard errors in parentheses. *Four loci analysed.

110 microsatellite markers were derived from either a tiger quoll or a mixed tiger/eastern quoll genetic library. The isolation of these markers and amplification procedures used in this study have been described elsewhere (Firestone 1999).

Aliquots of the PCR products were mixed with an equal volume of formamide loading dye, heated to 80°C, and loaded on a 6% polyacrylamide sequencing gel containing 50% w/v urea. Gels were fixed (10% glacial acetic acid/10% methanol) and dried, then exposed to autoradiographic film (X-OMAT, Kodak or Hyperfilm-HP, Amersham). Alleles were scored by comparison with a size marker (M13 sequence, USB) electrophoresed alongside the samples on each sequencing gel. Eighteen populations were typed at all six loci; ST was typed at four loci while AR was typed at five loci.

Statistical Analyses

Deviations from Hardy-Weinberg equilibrium (HWE) were tested in each population by one of two methods as implemented in GENEPOP (Raymond & Rousset 1995): either complete enumeration for loci with up to four alleles (Louis & Dempster 1987) or by a Markov chain method for loci with five or more alleles (Guo & Thompson 1992). Genetic variability of species and populations was measured as the number of alleles per locus (A) and unbiased

expected heterozygosity (HE) for each locus using BIOSYS (Swofford & Selander 1981).

Differences of mean HE between species and populations were tested by analysis of variance

(ANOVA; Sokol and Rohlf 1981) and post hoc hypotheses were examined using Scheffe’s test. Empirical studies have shown that population size and level of genetic variability are positively correlated (Frankham 1996). Similarly, measures of genetic diversity are likely to be affected by small

111 sample size (e.g. Roy et al. 1994); A more so than HE (Nei et al. 1975; Bouzat et al. 1998). Sample sizes in this study ranged from three to 58 individuals per population, therefore, the effects of limited sampling regimes in assessing allelic diversity could not be discounted. Thus, Monte Carlo simulations were used to calculate the expected number of alleles in an infinite population (e.g. Roy et al. 1994). To compare the number of alleles in species that differed in sample size, the expected number of alleles in an infinite population by Monte-Carlo simulations were calculated (Roy et al. 1994). Individuals were selected at random without replacement and the cumulative number of alleles were calculated until all individuals had been sampled. This procedure was repeated 1000 times for each species and the mean and standard deviation of the number of alleles was calculated as a function of sample size. A quasi-Newton best fit curve was then applied to the means using the equation y= a x/(x+ß) where y = number of alleles, and x = number of individuals. In this equation a and ß are constants, where a represents the number of alleles in an infinite population. In addition, I employed residual analysis as a measure of corrected allelic diversity (A’) using ANOVA and Scheffe’s post hoc tests, to examine interspecific and interpopulation differences in number of alleles. The data were log-transformed prior to the analysis to accommodate nonlinearity and deviation from the normal distribution. To control for the effect of sample size, I have used residuals generated by linear regression of sample size versus number of alleles.

Genetic differentiation was assessed by a number of methods. First, the number of unique or private alleles found between populations or species may be seen as a measure of genetic differentiation; however similarly to the number of alleles, the number of unique alleles is also affected by the extent of the sampling regime. When closely related species are compared, the number of unique alleles found within each species is a measure of genetic distinction. However, this is strongly influenced by the sample size and geographic scope of the sampling within each taxon. Thus the expected number of unique alleles for each species in comparison with another species, given different sample sizes, were calculatedusing Monte- Carlo simulations as above.

Second, I summarized allele frequencies for populations into two dimensions using multidimensional scaling (MDS), which makes few assumptions of the structure of the data. MDS analysis was performed using a convergence factor of 0.005 and 50 iterations as implemented in STATISTICA (Statsoft, Inc.). Finally, genetic distances and population subdivision among all pairwise comparisons of populations were estimated using methods based on both the step-wise mutations model (SMM) and the infinite alleles model (IAM), since neither mutation model is strictly

112 correct for microsatellites (Primmer et al. 1998). Thus Goldstein’s dm2 distance (SMM)

(Goldstein et al. 1995) and Nei’s unbiased genetic distance (Nei’s D; IAM) (Nei 1972) were

employed. Subdivision among populations was estimated by both RST (SMM) (Slatkin

1995; Goodman 1997) and FST (IAM) (Wright 1951). A Mantel procedure was used to test

2 for correlation both between dm and Nei’s D and between RST and FST. Nei’s D was further used to construct a neighbor-joining tree as implemented in the NEIGHBOR program in PHYLIP (Felsenstein 1995). Bootstrap analysis was done by first generating 1000 distance matrices using MICROSAT (Minch et al. 1998); 1000 bootstrapped neighbor- joining trees were then constructed using the NEIGHBOR program and summarized by the

CONSENSE program in PHYLIP (Felsenstein 1995). Significance of all pairwise FST values was assessed by 10,000 iterations as implemented in FSTAT (v. 2.8) (Goudet 1999).

Furthermore, Mantel tests were performed to assess the relationship of genetic differentiation

between populations (FST) to that of geographic distance.

Lastly, an assignment test (available from www.biology.ualberta.ca/jbrzusto/Doh.html) was performed to determine how characteristic an individual’s genotype was of both the species and population from which it was sampled (Paetkau et al. 1995; Paetkau et al. 1998). The expected frequency of each individuals’ genotype was calculated at both the species and population levels and assigned to the species or population for which the expected frequency was greatest. All frequencies were adjusted to avoid zeros using the method of Titterington et al. (1981).

113 Results

Genetic Variability of Microsatellites In Quolls

The six microsatellite loci used in this study were highly polymorphic in all species examined, with 14-23 total different alleles per locus (Appendix 5). Uncorrected mean A per population

ranged from 2.5 (MW, BS, and SB, tiger quolls) to 9.2 (BF, western quolls) and mean HE ranged from 0.469 (BS, tiger quolls) to 0.883 (PE, western quolls). The samples analysed in both GI and WY populations were monomorphic at a single locus each (locus 1.3 and 3.3.1, respectively); the samples analysed from the AT population were monomorphic at two loci,

1.3 and 4.4.10. Allele frequency distributions were highly skewed at each locus, generally with two or three common alleles and many rare alleles.

Deviations From Hardy Weinberg Equilibrium

Some loci deviated from HWE proportions in eleven of the twenty populations (see Appendix

7). All six loci in the BF population deviated from HWE. Deviations were also found in three loci from NE (1.3, 3.1.2, 4.4.2), and GL (3.1.2, 3.3.1, 3.3.2) populations; two loci from each of TE (3.3.2, 4.4.2), KA (3.1.2, 3.3.2) and PE (1.3, 3.3.1) populations; and one locus from each of the MW (1.3), BS (3.1.2), WY (4.4.2), ST (3.3.1), and VB (3.3.2) populations. No locus was prevalent in deviations from HWE.

I also examined overall HW proportions, combined over all loci, for each population. Of all tiger quoll populations, only one (WY) was significantly deviant from overall HWE at all six loci (c2 = 24.2, P = 0.007); this was due to a general heterozygote deficiency. Among eastern quoll populations NE, ST, GL and VB were all significantly different from genotype proportions expected under HWE, over all loci. Significant deviation of GL (c2 infinity, P highly significant) was due to heterozygote excesses at two loci (3.3.1, 3.3.2) and

114 heterozygote deficits at one locus (3.1.2). All other significant differences in eastern quolls were due to heterozygote deficits (NE, c2 infinity, P highly significant; ST, c2 = 19.9, P =

0.0029; VB, c2 = 28.2, P = 0.0052). Among northern quolls, KA was the only population significantly different from overall HWE proportions (c2 = 35.3, P = 0.0004); this deviation was due to heterozygote deficits. In addition, neither western quoll population was in HWE proportions. Overall differences from HWE in the BF and PE populations (BF, c2 = 78.2, P

< 0.0001; PE, c2 infinity, P highly significant) were due to heterozygote deficiencies.

Genetic Diversity Among Species

Monte Carlo simulations of the estimated cumulative alleles for each species are shown in

Figure 15. The cumulative number of alleles begins to asymptote between 10-20 samples for most species except for D. viverrinus, which begins to asymptote after approximately 25 individuals are sampled. These simulations indicate that our sampling regime was adequate in picking up a substantial proportion of alleles present for most populations.

Uncorrected allelic diversity (A) is shown in Table 6. In general, western quolls had higher numbers of alleles than any other species, while tiger quolls had lower allelic diversity than other species. Measures of genetic diversity after correction for sample size differences are shown in Figure 16. Significant differences were found among species in both corrected

allelic diversity (A’; ANOVA F = 10.59; P £ 0.0001 ) and mean HE (ANOVA F = 5.70, P

= 0.001) (Figure 16A). Post hoc tests revealed that HE

115 Figure 15. Estimated cumulative number of alleles per sample size for each species by Monte Carlo simulation. Curves were fitted using the equation y = a x/(x+ß) where y = number of alleles and x = number of individuals. In these equations a represents the number of alleles in an infinite population. A) D. maculatus, a = 57.0, ß = 10.87, r2 = 0.979; B) D. viverrinus, a = 69.70, ß = 20.88, r2 = 0.965; C) D. hallucatus, a = 76.11, ß = 9.54, r2 = 0.999; D) D. geoffroii, a = 80.23, ß = 10.61, r2 = 0.993. Note that y-axis scales are different between species.

A B 80.0 Dv 60.0 Dm 60.0 50.0

40.0 40.0 30.0

20.0 20.0 Mean number of alleles ± s.d. Mean number of alleles ± s.d. 10.0

0.0 0.0 0 25 50 75 100 125 0 25 50 75 100 125 150 Number of individuals selected Number of individuals selected

C D 70.0 Dh 80.0 Dg 60.0

50.0 60.0

40.0 40.0 30.0

20.0

Mean number of alleles ± s.d. 20.0

10.0 Mean number of alleles ± s.d.

0.0 0.0 0 5 10 15 20 25 30 35 40 45 0 10 20 30 40 50 60 Number of individuals selected Number of individuals selected

116 Figure 16. Mean and standard error of corrected allelic diversity (A’) and expected heterozygosity (HE) at six microsatellite loci (A) among species and (B) within species of quolls. Horizontal bars indicate significant differences among groups by Scheffe’s post hoc test. Dm = D. maculatus, Dv = D. viverrinus, Dh = D. hallucatus, Dg = D. geoffroii. See Table 6 for key to population names.

A

0.4 1.5 0.3

0.2 A’ 1.0 0.1 Mean HE 0.0 0.5 -0.1

-0.2 Dm Dv Dh Dg 0.0 Dm Dv Dh Dg Species Species

B

0.6 1.0 0.4

0.2

A’ 0.0 Mean HE

-0.2 0.5

-0.4

-0.6

-0.8 MW TT BT BS GI GL TE ST AT PE 0.0 BG CH CO WY SB VB NE KA AR BF CH CO WY BS MW NE VB GL AR BF BG BT TT GI SB ST TE KA AT PE Dm Dv Dh Dg Dm Dv Dh Dg Population Population

117 was significantly higher in populations of western quolls than in any other species; western quolls also had a significantly higher allelic diversity than either tiger quolls or eastern quolls, but were not significantly different to northern quolls. Eastern, tiger, and northern quolls were

not significantly different from each other in either A’ or mean HE.

Genetic Diversity Among Populations Within Species

When each species was examined separately, significant differences in A’ were found among populations of all species except western quolls (Figure 16B). Among tiger quoll populations,

GI and SB had significantly lower A’ than MW, BG, TT, CH, and BT (ANOVA F = 9.83, P

< 0.0001). Among eastern quoll populations, GL had significantly higher allelic diversity than

TE, NE, and ST; furthermore, ST had significantly lower allelic diversity than GL, VB, and TE

(ANOVA F = 14.18, P < 0.0001). Among northern quoll populations, KA had a significantly higher number of alleles than either AR or AT (ANOVA F = 11.01, P < 0.001).

There were no differences in allelic diversity between the two western quoll populations

(unpaired t-test, t = 0.31; P = 0.76).

Significant differences in levels of HE also were found among eastern quoll populations

(ANOVA F = 2.88, P = 0.045) and western quoll populations (ANOVA F = 8.77, P =

0.014), however post hoc tests of eastern quolls indicated that there were no significant

pairwise differences. Among western quoll populations, PE had significantly higher HE than

BF (Figure 16B).

118 Genetic Differentiation Among Species: Unique Alleles

Each species possessed only a subset of the total alleles found in the genus (Appendix 5).

While there was great overlap among species in the alleles present (e.g. alleles 101-107 were found in all four species at locus 3.3.1), there was also substantial partitioning at each locus

(e.g. at locus 3.3.1, alleles 91, 93, 97, and 117 were unique to northern quolls, whereas alleles 127-145 were found only in western quolls). The effect of our sampling regime on the number of unique alleles observed among species was examined and the proportion of unique alleles estimated by Monte Carlo simulations is shown in Figure 17 and Table 7. In general

D. geoffroii had the highest number of unique alleles in relation to all other species, whereas

D. maculatus had the lowest number of unique alleles in relation to all other species. After approximately 30-40 individuals of most species had been sampled the graphs begin to asymptote (e.g. Figure 17A, C, D) indicating that the proportion of unique alleles found in relation to other species has peaked. D. viverrinus, however, may require additional samples to reach this asymptote (Figure 17B).

Genetic Differentiation Among Species: Allele Frequencies

Genetic differentiation was also examined using allele frequency data. The differences in allele frequencies among species and populations of quolls were summarized using MDS (Figure

18) which showed that populations within species generally clustered closely in two- dimensional space. Furthermore, these data fit in two dimensions with low stress (0.10) and with a high proportion of the variance accounted for (r2 = 0.94), indicating a good fit of the data in multidimensional space.

119 Figure 17. Estimated number of unique alleles for each species by Monte Carlo simulations. Curves were fitted using the equation y = a x/(x+ß) where y = number of alleles and x = number of individuals. In these equations a represents the number of unique alleles in an infinite population. A) No. of unique alleles in Dm compared with Dg [a = 22.32, ß = 18.51, r2 = 0.98], Dh [a = 21.22, ß = 14.93, r2 = 0.96], and Dv [a = 16.63, ß = 26.44, r2 = 0.99]. B) No. of unique alleles in Dv compared with Dm [a = 25.22, ß = 20.74, r2 = 0.95], Dh [a = 26.68, ß = 35.72, r2 = 0.97], and Dg [a = 28.86, ß = 48.22, r2 = 97]. C) No. of unique alleles in Dh compared with Dm [a = 35.02, ß = 11.09, r2 = 0.99], Dg [a = 30.70, ß = 8.32, r2 = 0.99], and Dv [a = 25.46, ß = 9.70, r2 = 0.99]. D) No. of unique alleles in Dg compared with Dm [a = 42.05, ß = 12.19, r2 = 0.99], Dh [a = 38.14, ß = 9.40, r2 = 0.99], and Dv [a = 34.36, ß = 12.64, r2 = 0.99]. Note that y-axis scales are different between species.

A B

25 Dv 25 Dm Dm Dh s.d.

± 20 Dg 20 Dg ± s.d. Dh 15 15 Dv 10 10

5 5

0 0 0 25 50 75 100 125 150 Mean number of unique alleles

0 25 50 75 100 125 Mean number of unique alleles Number of individuals selected Number of individuals selected

C D Dh 30 40 Dg Dm Dm

s.d. Dg 25 s.d. ± Dh ± 30 20 Dv Dv

15 20

10 10 5

0 0 0 10 20 30 40 50 0 10 20 30 40 50 60 Mean number of unique alleles Mean number of unique alleles Number of individuals selected Number of individuals selected

120 Table 7. The number and percentage (in parenthesis) of unique alleles between four species of quolls.

Dm Dv Dh Dg

Dm (54) *** 14 (25.9) 20 (37.0) 20 (37.0) Dv (63) 23 (36.5) *** 22 (34.9) 22 (34.9) Dh (62) 28 (45.2) 21 (33.9) *** 26 (41.9) Dg (70) 36 (51.4) 29 (41.4) 34 (48.6) *** Total number of alleles in each species is in parenthesis beside the species name at the left column of the table. All 20 populations were included.

Figure 18. MDS based upon Euclidean distances among 15 populations of quolls. (l) D. maculatus, (9) D. viverrinus, (n) D. hallucatus, (s) D. geoffroii. Stress = 0.10, Rsq = 0.94.

1.4 GL : 1.0 TE GI : VB : 0.6

SB 0.2 BG BS BT CH -0.2 CO BF KA TT s Dimension 2 n AT -0.6 n

-1.0 MW

-1.4 -1.4 -0.8 -0.2 0.4 1.0 1.6 2.2 Dimension 1

121 Genetic Differentiation Among Populations Within Species: Population Subdivision

Pairwise genetic differentiation measures among populations were also estimated by both FST and RST. A Mantel test showed that FST and RST values were highly correlated (r = 0.72; P

£ 0.001); therefore only FST values are shown in Table 8. However pairwise values of Fst involving GL or TE populations were consitently in disagreement with Rst values between the

same pairs of populations. All pairwise comparisons of FST values were tested for significance. Most pairwise comparisons showed significant population subdivision both between populations and between species (Table 8). An exception was found among populations of tiger quolls, particularly among the four populations from the Barrington area.

None of these pairs of populations showed significant subdivision. Additionally, the SB population was not differentiated from most other populations, but this is probably due to the very low sample size for this population.

Genetic Differentiation Among Populations Within Species: Genetic Distance

Measures

A Mantel test of Nei’s D and Goldstein’s dm2 indicated that these distance measures were significantly correlated (r = 0.64; P £ 0.001). I therefore present pairwise values of Nei’s D only (Table 8). Examination of both genetic distances among tiger quoll populations revealed that MW was consistently the most distant from all other tiger quoll populations. Lowest genetic distances were found among the four geographically close populations of tiger quolls

(CO, CH, BG, BT) by Nei’s D; the results for dm2 were not consistent with Nei’s D, however (data not shown). Among eastern quolls, both Nei’s D and dm2 values indicated that the Tasmanian populations

122 - 0.256* 0.272* 0.250* 0.266* DG - AT BF 0.472 0.462 0.432 0.223* 0.225* - KA 0.67* DH 0.329* 0.329* 0.399* 0.269* - TE 0.356 0.078* 0.323* 0.385* 0.273* - VB 0.222* 0.109* 0.384* 0.440* 0.329* - GL 0.411*0.416* 0.347* 0.350* 0.349* 0.340* 0.425* 0.221* DV (above the diagonal). - TT ST 0.222 0.327* 0.377*0.156* 0.342* 0.376*0.156* 0.351*0.128* 0.319* 0.356* 0.361* 0.257* 0.393* 0.312* 0.307* 0.441* 0.332* 0.309* 0.299* 0.390* 0.331* 0.374* 0.299* 0.396* 0.323* 0.174* 0.370* 0.370* 0.394* 0.185* 0.196* - 0.136 0.086 0.110 0.117 - BS SB 0.310 0.207 0.167*0.117* 0.184*0.143* 0.125*0.124* 0.393* 0.327* 0.322* 0.313* 0.385* 0.161* - BT 0.058 - BG 0.029 0.019 - CH 0.011 0.047 0.000 0.183* 0.169* 0.218* - CO 0.242 - GI 0.445* 0.156* 0.221* 0.221* 0.275* 0.399* 0.429* 0.304* 0.451* 0.385* 0.389* 0.378* 0.486* 0.242* - MW 1.59 0.270.48 0.64 0.51 0.420.64 0.391.01 0.03 0.461.76 0.11 0.600.69 0.01 0.36 0.07 0.32 0.97 0.03 0.501.80 0.20 0.251.50 0.12 0.33 0.971.40 0.27 0.36 1.19 0.20 1.66 1.17 0.20 0.38 1.632.83 0.23 1.513.53 1.46 1.57 0.26 3.08 0.16 1.31 2.92 1.391.04 1.44 0.53 3.00 1.52 2.54 1.29 1.65 0.54 1.55 2.99 1.54 2.48 1.43 0.81 1.64 2.47 1.46 2.65 1.53 0.97 1.59 2.88 1.71 2.28 1.56 1.11 1.56 2.69 1.98 1.43 1.05 0.41 2.83 2.38 0.14 1.65 2.91 2.93 0.15 2.73 2.12 1.76 1.20 2.03 1.82 2.21 2.12 1.90 2.51 2.28 2.08 1.93 DM Table 8. Pairwise comparison of Nei’s D (below the diagonal) and F SPECIES/POPULATION TIGER QUOLL MW GI CO CH BG BT BS SB TT EASTERN QUOLL GL VB TE NORTHERN QUOLL KA AT WESTERN QUOLL BF Boxes denote major groups. Asterisks indicate significant differentiation between pairs of populations after Bonferroni correction for multiple comparisons.

123 (GL, VB, TE) were closer to each other than to the mainland populations (NE and ST; data not shown). Nei’s D also showed that the two northern quoll populations (KA and AT) were more closely related to each other than either was to any other population.

Nei’s D was used to build a neighbour joining tree among 15 populations of quolls for which reliable data were available (Figure 19). These data show that all species form their own clades, although bootstrap support is very low in many cases. Similar to trees based on mtDNA sequences, northern quoll populations are the most distant from other populations based on microsatellites and thus form an early split. The topology of this tree, however, is not consistent with those based on mitochondrial DNA loci (Krajewski et al. 1997; Firestone in press).

Geographic distance vs. FST

The relationship of genetic subdivision (FST) to geographic distance was examined by a

Mantel test within each of the three species in which more than two populations were

available. There was a significant correlation between geographic distance and FST for tiger quoll populations (r = 0.615; P = 0.01), indicating that distance explains a substantial amount of the genetic variance observed between populations of this species. This correlation did not hold for either eastern quolls (r = 0.603; P = 0.06) or northern quolls (r = 0.837; P = 0.16).

It should be noted, however, that the range of geographic distances and the number of populations available were low in these two species in relation to that of tiger quolls.

124 Figure 19. Neighbor-joining tree of 15 populations of quolls using Nei’s D. Numbers at nodes indicate the percent of trees from 1000 bootstrap replicates that have the branch to the right.

37 GI 47 0.5 SB 28 BS

16 BG Tiger quolls 21 CH

20 BT 9 TT 46 CO 49 MW BF Western quoll 89 61 VB 87 TE Eastern quolls GL AT Northern quolls KA

125 Assignment tests

Results from the assignment test show that 252 (98.4%) individuals were correctly assigned to their true species (Table 9). Only 4 animals (1.6%) were misassigned at the species level, each of which was only partially genotyped: one D. maculatus individual was misassigned as

D. geoffroii; two D. viverrinus were misassigned as either D. maculatus or D. hallucatus; and one D. geoffroii was misassigned as D. maculatus. Results of population assignments showed that 211 (82.4%) animals were correctly assigned to their source populations (Table

10). Of the animals that were misassigned, 33 (12.9%) were misassigned to a geographically close population; 8 (3.1%) were misassigned to more distant populations within the same species and 4 animals (1.6%) were misassigned to a different species. Three of the four individuals misassigned to different species at the population level were the same as those misassigned at the species level (Table 9).

Table 9. Species assignment test.

Assigned Species % success

Source Dm Dv Dh Dg Species Dm (96) 95 - - 1 99 Dv (93) 1 91 1 - 98 Dh (32) - - 32 - 100 Dg (35) 1 - - 34 97

Individuals were assigned to the species from which their genotypes were most likely to occur. Number of assignments from species i (row) to species j (column).

126 % 80 11 76 68 83 75 91 91 95 64 83 100 100 success 1 27 - 33 94 AT BF -- Dh Dg KA ------26 1 5 TE 4 VB - 20 1 Dv GL ------53 3 2 9 TT ------2 ------10 1 ------2 -- BS SB ------5 - 1 BT 3 1 ------BG 1 ------CH 1 ------CO - 11 42 13 14 3 21 3 1 15 1 1 1 1 GI ------4 Dm Assigned Population MW 12 ------Source Pop MW (12) GI (5) CO (9) CH (17) BG (22) BT (11) BS (6) SB (3) TT (11) GL (58) VB (21) TE (14) KA (26) AT (6) BF (35) Table 10. Population assignment test for 15 populations of quolls. Individuals were assigned to the populations from which their genotypes most likely occur. Number of assignments from population i (row) to j (column) are indicated. Boxes denote populations within species that are in close geographic proximity.

127 Discussion The microsatellites used here were extremely useful for assessing variability and differentiation at all levels (within populations, among populations within species, and among species). Quolls generally exhibited levels of variability within the range of that found in other species, although western quolls had higher heterozygosity than that reported from some other species (e.g. Roy et al. 1994; Houlden et al. 1996; O'Ryan et al. 1998).

Hardy-Weinberg Equilibrium A number of populations were not in Hardy-Weinberg equilibrium, suggesting that null alleles may be present. It appears as if some eastern quoll populations and both western quoll populations may have null alleles. An alternative explanation for these populations not exhibiting HWE may be the Wahlund effect, which leads to heterozygote deficits. The Wahlund effect is the result of very recent population mixing (i.e. in the current generation, or at the time of sampling). It is possible that population mixing may result in the phenomena of both high levels of allelic diversity as well as decreased levels of heterozygosity. Recent population mixing may be the case in western quolls where both high levels of diversity and low levels of heterozygosity exist. On the other hand, eastern quolls show no evidence of abnormally high diversity, so null alleles may be a better explanation. Most eastern quoll samples were taken from specimens held in museums; damage to DNA through the tanning process or over long periods of time can lead to mis-priming and null alleles.

Genetic Diversity Among And Within Species Genetic diversity was significantly different among the four Australian quoll species and was not as expected on the basis of either primer origin or population history. Higher levels of allelic diversity and heterozygosity were found among western quolls than any other species. Western quolls had more than twice the mean number of alleles than tiger quolls, yet only half the number of individuals were sampled. This high level of diversity among western quolls was particularly surprising given the widespread and long-term decline of this species. This finding is also notable in light of the microsatellite primers having been designed from tiger quolls and eastern quolls. There is a possibility that mutations arising in the priming site for microsatellites may lead to the presence of null alleles, and thereby, lower levels of heterozygosity in species where heterologous primers are employed. It has been shown previously that polymorphism at microsatellite loci may decline with increasing phylogenetic distance from the species for which the primers were originally characterized (Moore et al. 1991; FitzSimmons et al. 1995). However, this was not the case with western quolls. The high levels of genetic diversity within the BF population of western quolls may be an artefact of the

128 captive breeding program although the historic population (PE) also had high genetic diversity. Alternatively the high levels of variation in western quolls could reflect past large population sizes. Given the relatively recent decline of this species, there may be a considerable lag before diversity is lost. Maintaining these high levels of diversity in the BF population is important in light of the intensive captive breeding and reintroduction program instigated to assist conservation of this species. In addition, high levels of microsatellite variability may be useful for tracking paternity and reproductive success within that colony. Northern quolls, tiger quolls and eastern quolls were not significantly differentiated from one another in either measure of diversity, indicating that the same genetic processes may be operating within these three species.

Differences in variability within species were also apparent and were as expected (i.e. low variation was found in small or isolated populations). The GI and SB populations of tiger quolls had lower corrected allelic diversity than that of either the MW, BG, TT, CH, or BT populations. Both the GI and SB populations are small: GI has been a captive bred colony for approximately 18 years with little genetic exchange (Bruce Kubbere, pers. comm.) and the

SB population is from an area in Victoria where there have been very few sightings or records over the last decade. In contrast, the MW, BG, CH, and BT populations are wild populations with relatively high numbers of individuals. Similarly, there were differences in allelic diversity among populations of eastern quolls and among populations of northern quolls.

The mainland populations of eastern quolls (NE, ST) had significantly lower allelic diversity than populations in Tasmania (GL, VB, TE). This result might be surprising given that theory predicts low genetic diversity among island populations when compared to mainland populations (Frankham, 1997). Two possible explanations exist: lower levels of diversity may be an artefact of the difficulties in amplifying DNA from museum tissues, or the mainland

129 populations may have been on the brink of extinction when these samples were collected.

Within northern quolls, the KA population had higher allelic diversity than either AT or AR.

Again, the KA population is in a relatively stable state, with large population numbers extended over a wide area. It is thought that the AT population is now isolated and in decline and the AR population is extinct.

Genetic Differentiation Among and Within Species

Each species possessed some unique alleles (Table 7), which were useful in defining species clusters. Phylogenetic analysis, based on distance values among populations, (Nei’s D, Table 8; Figure 19) was able to partition species into distinct clades, although bootstrap support was limited. In phylogenetic reconstructions based on genetic distances between mtDNA sequences, eastern and western quolls are sister species (Krajewski et al. 1997; Firestone in press); in the reconstruction based on distances between microsatellite alleles, tiger and western quolls are sister species (Figure 19). At deeper nodes, phylogenetic reconstructions based on more recent markers such as microsatellites are less reliable than those based on more distant markers such as mtDNA. This is due to the extremely high mutation rate of microsatellites and homoplasy of alleles (similarity in phenotype, but not identity by decent) (Estoup et al. 1995). MtDNA phylogenies provide better resolution at deeper nodes whereas microsatellites may provide greater resolution at the terminal nodes. For instance, among eastern quoll populations, the phylogeny based on microsatellites shows more structure than the phylogeny based on mtDNA.

Similar to microsatellite phylogenetic reconstruction, however, MDS analysis was able to partition populations into species clusters based on the presence of unique alleles and differences in allele frequencies (Figure 18). The positioning of populations into distinct species clusters indicates that the microsatellites used here are potentially useful for identifying different species of quolls in forensic tests. The assignment test (Table 9) and analysis of the mitochondrial DNA control region (Firestone in press) also may serve this function and may be even more useful when individuals are considered.

130 Most population pairs within species were significantly differentiated from one another based

on allele frequencies (pairwise FST values; Table 8). This indicates that most populations should be considered as separate management units (MUs) according to recommendations by

Moritz (1994). One notable exception is found among tiger quoll populations. The four populations from the Barrington Tops region (CO, CH, BG, BT) are all located within a radius of 50 km, and were not significantly subdivided based on microsatellite loci; similarly, there were difficulties in assigning individuals correctly among these populations (Table 10).

However, studies of allele frequencies of the mtDNA control region have shown that some of these populations are actually differentiated (Firestone et al. 1999). Another exception may be found among many of the pairwise comparisons of SB; the lack of genetic differentiation between SB and these other populations may be due to the very small sample size of this population; the same may hold true for the GI and AT populations (Table 8).

Conservation Implications

Western quolls. Western quolls possessed greater allelic diversity and levels of heterozygosity than the other species. In addition, western quolls also possessed the greatest number of unique alleles in relation to other species. The recovery plan for western quolls was begun in 1991 with the breeding colony consisting of 20 captive founders (3 males, 5 females, and 12 young from two litters) and additional wild caught young used to augment the captive population (Serena et al. 1991). Supplemental wild-caught males have been periodically introduced to the captive colony for breeding purposes, and surplus young have been routinely released to one of several different unoccupied translocation sites (Serena et al. 1991; Orell & Morris 1994). Maintaining high levels of genetic variability within the

131 captive colony of western quolls is important to the long term viability of the translocated wild populations. Current estimates of variability in the extant population (BF) compared to a historic population (PE) show no differences in the mean number of alleles but higher levels of heterozygosity in the extant population. These high levels of diversity and heterozygosity may be a manifestation of the interbreeding of two populations, since the BF population is made up of captive and wild caught animals and the PE population, while comprised of individuals from one locality, were sampled over an extended period.

Breeding programs may greatly influence the levels of diversity within a captive population and the results presented here could indicate that the captive breeding program has been successful in maintaining high levels of diversity. Due to the success of the recovery program, including wide spread fox baiting, this species has been downgraded from ‘endangered’ to

‘vulnerable’ by the IUCN. Continued genetic monitoring of the captive and translocated populations is recommended as a means of assessing inbreeding or founder effects in recolonized areas.

Tiger quolls. Tiger quolls had low numbers of alleles in comparison with other species, however when this was corrected for sample size, there was no significant difference between

tiger, eastern, or northern quolls in levels of HE or in A’, indicating that the same evolutionary forces may be operating on these species.

Genetic subdivision shows that many populations of tiger quolls are separate MUs for

conservation purposes (FST values; Table 8) although microsatellite data did not show subdivision amongst the populations from the Barrington Tops region (CH, CO, BG,

132 BT). However, other studies have shown that the Barrington region populations are actually subdivided based on frequency differences in mtDNA, which is likely due to sex biased dispersal in these populations (Firestone et al. 1999).

In addition, the MW population (ascribed to D. maculatus gracilis) has been shown to be part of the mainland ESU, but the TT population forms a separate ESU to that on the mainland (Firestone et al. 1999). The TT population should be managed as a separate taxon to all other populations, whereas translocations between different MUs belonging to the same

ESU may be advisable in cases where population numbers have dropped to low levels.

However, analysis of FST versus geographic distance was significant for this species, indicating that distance itself explains much of the genetic variance observed between populations of tiger quolls. This implies that these animals are quite stationary, and dispersal distances are rather short relative to the large distances between sample localities. For this reason alone, it is important not to mix populations by reintroducing individuals from one site to another unless they are in close proximity.

Eastern quolls. Among the populations sampled, eastern quolls from Tasmania (GL, VB,

TE) had higher allelic diversity than those from the mainland (NE, ST; Figure 16).

Furthermore, significant differentiation exists between the three Tasmanian populations of

eastern quolls at the microsatellite loci examined (pairwise FST values; Table 8) indicating that each of these populations should be considered as separate MUs.

Eastern quolls from the mainland are currently presumed extinct; the last confirmed sighting of a mainland eastern quoll was in 1963 (Australian Museum records). The

133 reintroduction of eastern quolls from Tasmania to the mainland has been proposed in the past. However, there are still occasional reported sightings of eastern quolls from various mainland sites, and previously ‘extinct’ species have been resurrected in the past (e.g. Sinclair et al. 1996). If remnant populations of eastern quolls do still exist in remote areas of the mainland, then mixing of potentially different genetic units could prove to be deleterious.

Although extinct populations from the mainland were analysed, no conclusions can be drawn regarding differentiation of mainland and island populations due to missing data. Thus it is not clear whether relocations from Tasmania to the mainland would alter or reduce genetic variation of a remnant population, but it would be wise for such relocations to be postponed until the species is no longer just ‘presumed’ extinct on the mainland.

Northern quolls. In the past, four subspecies of D. hallucatus were recognized on the basis of morphological differences and geographical location (Gould, 1842; Thomas, 1909;

Thomas, 1926): D. h. hallucatus (Northern Territory; including the KA population), D. h. nesaeus (Groote Eylandt), D. h. exilis (Western Australia), and D. h. predator (Cape York

Peninsula, Queensland; including the AR and AT populations). However these trinomials are no longer in current use (e.g. Strahan, 1998; Maxwell et al., 1996). Furthermore, the lack of taxonomic clarity has proven to be a major stumbling block in the conservation of many species (e.g. Daugherty et al., 1990; Zink and Kale, 1995).

Preliminary genetic analysis of northern quolls has shown that the KA and AT populations are separate MUs based on significant differences in allele frequencies

134 (Table 8); no conclusions about population subdivision of the AR population could be drawn however, due to missing data. Other studies examining mtDNA (Firestone in press) have shown that there are two reciprocally monophyletic clades within northern quolls (Northern

Territory versus Queensland clades) corresponding to separations between the KA and AT populations in microsatellite allele frequencies shown here. Preliminary results suggest that these two populations may thus represent two distinct ESUs as well as different MUs. The detection of distinct ESUs implies historic separation and divergence between groups; therefore their separate management is recommended, to allow for continued divergence and evolution of the ESUs. The KA population from the Northern Territory should be recognized as a distinct conservation unit separate to the AT population from Queensland. The taxonomy should reflect this and the subspecific designations for these two groups should be resurrected.

I was able to examine only a few populations of northern quolls, however, and only a few individuals from two of those populations. I plan further studies to include the remaining geographically disjunct populations of northern quolls and additional markers (e.g. the mtDNA control region) to more thoroughly assess diversity and differentiation within this species.

In conclusion, the use of microsatellite markers has proven to be very effective in determining both levels of genetic variability and the degree of differentiation amongst all Australian species of quolls. Results presented here provide a genetic basis for future population monitoring and should prove useful to conservation managers and agencies in decision making processes related to the conservation of these species.

135 References

Allendorf FW, Leary RF (1986) Heterozygosity and fitness in natural populations of animals.

In: Conservation Biology: The Science of Scarcity and Diversity. (ed. Soulé ME), pp. 57-

76. Sinauer Associates, Sunderland, Massachusetts.

Boom R, Sol CJA, Salimans MMM, Jansen CL, Wertheim-van Dillen PME, van der

Noordaa J (1990) Rapid and simple method for purification of nucleic acids. Journal of

Clinical Microbiology, 28, 495-503.

Bouzat JL, Cheng HH, Lewin HA, Westemeier RL, Brawn JD, Paige KN (1998) Genetic evaluation of a demographic bottleneck in the greater prairie chicken. Conservation Biology,

12, 836-843.

Braithwaite RW, Griffiths AD (1994) Demographic variation and range contraction in the northern quoll, Dasyurus hallucatus (Marsupialia:Dasyuridae). Wildlife Research, 21, 203-

217.

Brunner PC, Douglas MR, Bernatchez L (1998) Microsatellite and mitochondrial DNA assessment of population structure and stocking effects in Arctic charr Salvelinus alpinus

(Teleostei: Salmonidae) from central Alpine lakes. Molecular Ecology, 7, 209-223.

136 Crozier RH (1992) Genetic diversity and the agony of choice. Biological Conservation, 61,

11-15.

Daugherty, CH, Cree, A, Hay, JM, and Thompson, MB (1990). Neglected taxonomy and continuing extinctions of tuatara (Sphenodon). Nature, 347, 177-179.

Estoup, A, Tailliez, C, Cornuet, J-M, Solignac, M (1995). Size homoplasy and mutational processes of interrupted microsatellites in two bee species, Apis mellifera and Bombas terrestris (Apidae). Molecular Biology and Evolution, 12, 1074-1084.

Estoup A, Rousset F, Michalakis Y, Cornuet J-M, Adriamanga M, Guyomard R (1998)

Comparative analysis of microsatellite and allozyme markers: a case study investigating microgeographic differentiation in brown trout (Salmo trutta). Molecular Ecology, 7, 339-

353.

Felsenstein J (1993) PHYLIP (Phylogeny Inference Package) Version 3.5c. Distributed by the author. Department of Genetics, University of Washington, Seattle.

Firestone K (1999) Isolation and characterization of microsatellites from carnivorous marsupials (Dasyuridae:Marsupialia). Molecular Ecology, 8, 1084-1086.

137 Firestone K, Elphinstone M, Sherwin B, Houlden B (1999) Phylogeographical population structure of tiger quolls Dasyurus maculatus (Dasyuridae:Marsupialia), an endangered carnivorous marsupial. Molecular Ecology, 8, 1613-1626.

Firestone KB (in press) Phylogenetic relationships among quolls revisited: the mtDNA control region as a useful tool. Journal of Mammalian Evolution.

FitzSimmons NN, Moritz C, Moore SS (1995) Conservation and dynamics of microsatellite loci over 300 million years of marine turtle evolution. Molecular Biology and Evolution, 12,

432-440.

Frankham R (1995) Inbreeding and extinction: a threshold effect. Conservation Biology, 9,

792-799.

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.

Gall GAE (1987) Inbreeding. In: Population Genetics & Fishery Management. (eds.

Ryman, N and Utter, F), pp. 47-87. University of Washington Press, Seattle.

138 García-Moreno J, Matocq MD, Roy MS, Geffen E, Wayne RK (1996) Relationships and genetic purity of the endangered Mexican wolf based on analysis of microsatellite loci.

Conservation Biology, 10, 376-389.

Goldstein DB, Linares AR, Cavalli-Sforza LL, Feldman MW (1995) An evaluation of genetic distances for use with microsatellite loci. Genetics, 139, 463-471.

Goodman SJ (1997) Rst Calc: a collection of computer programs for calculating estimates of genetic differentiation from microsatellite data and determining their significance. Molecular

Ecology, 6, 881-885.

Goodman SJ (1998) Patterns of extensive genetic differentiation and variation among

European harbour seals (Phoca vitulina vitulina) revealed using microsatellite DNA polymorphisms. Molecular Biology and Evolution, 15, 104-118.

Goudet J (1999) FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.8).

Gould J (1842) Characters of a new species of Perameles, and a new species of Dasyurus.

Proceedings of the Zoological Society of London, 1842, 41-42.

Guo SW, Thompson EA (1992) Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics, 48, 361-372.

139 Hoss M, Pääbo S (1993) DNA extraction from pleistocene bones by a silica-based purification method. Nucleic Acids Research, 21, 3913-3914.

Houlden BA, England PR, Taylor AC, Greville WD, Sherwin WB (1996) Low genetic variability of the koala Phascolarctos cinereus in south-eastern Australia following a severe population bottleneck. Molecular Ecology, 5, 269-281.

Jones ME (1995) Guild structure of the large marsupial carnivores in Tasmania. PhD

Thesis, University of Tasmania, Hobart, Australia.

Krajewski C, Young J, Buckley L, Woolley PA, Westerman M (1997) Reconstructing the evolutionary radiation of Dasyurine marsupials with cytochrome b, 12S rRNA, and protamine

P1 gene trees. Journal of Mammalian Evolution, 4, 217-236.

Kumar S, Tamura K, Nei M (1993) MEGA: Molecular Evolutionary Genetics Analysis.

The Pennsylvania State University.

Lacy RC (1997) Importance of genetic variation to the viability of mammalian populations.

Journal of Mammalogy, 78, 320-335.

Louis EJ, Dempster ER (1987) An exact test for Hardy-Weinberg and multiple alleles.

Biometrics, 43, 805-811.

140 Mansergh I (1984) The status, distribution and abundance of Dasyurus maculatus (tiger quoll) in Australia, with particular reference to Victoria. Australian Zoologist, 21, 109-122.

Maxwell S, Burbidge AA , Morris K, Eds. (1996) The 1996 Action Plan for Australian

Marsupials and Monotremes. Wildlife Australia, Canberra.

Minch E, Ruiz-Linares A, Goldstein D, Feldman M., Cavalli-Sforza LL (1998) Microsat

(version 1.5e): a computer program for calculating various statistics on microsatellite allele data. Stanford University, Stanford, CA. www: http://lotka.stanford.edu/ microsat.html

Moritz C (1994) Defining 'Evolutionarily Significant Units' for conservation. Trends in

Ecology and Evolution, 9, 373-375.

Moritz C, Lavery S, Slade R (1995) Using allele frequency and phylogeny to define units for conservation and management. American Fisheries Society Symposium, 17, 249-262.

Moore S, Sargeant L, King T, Mattick J, Georges M, Hetzel D (1991) The conservation of dinucleotide microsatellites among mammalian genomes allows the use of heterologous PCR primer pairs in closely related species. Genomics, 10, 654-660.

141 Nei M (1972) Genetic distance between populations. American Naturalist, 106, 283-292.

Nei M, Maruyama T, Chakraborty R (1975) The bottleneck effect and genetic variability in populations. Evolution, 29, 1-10.

Newman D, Pilson D (1997) Increased probability of extinction due to decreased genetic effective population size: experimental populations of Clarkia pulchella. Evolution, 51, 354-

362.

Orell P, Morris K (1994) Chuditch recovery plan 1992-2001. Como, WA, Department of

Conservation and Land Management.

O'Ryan C, Harley EH, Bruford MW, Beaumont M, Wayne RK, Cherry MI (1998)

Microsatellite analysis of genetic diversity in fragmented South African buffalo populations.

Animal Conservation, 1, 85-94.

Paetkau D, Calvert W, Stirling I, Strobeck C (1995) Microsatellite analysis of population structure in Canadian polar bears. Molecular Ecology, 4, 347-354.

Paetkau D, Shields GF, Strobeck C (1998) Gene flow between insular, coastal and interior populations of brown bears in Alaska. Molecular Ecology, 7, 1283-1292.

142 Paetkau D, Strobeck C (1994) Microsatellite analysis of genetic variation in black bear populations. Molecular Ecology, 3, 489-495.

Pope LC, Sharp A, Moritz C (1996) Population structure of the yellow-footed rock-wallaby

Petrogale xanthopus (Gray, 1854) inferred from mtDNA sequences and microsatellite loci.

Molecular Ecology, 5, 629-640.

Primmer CR, Saino N, Møller AP, Ellegren H (1998) Unravelling the processes of microsatellite evolution through analysis of germ line mutations in barn swallows Hirundo rustica. Molecular Biology and Evolution, 15, 1047-1054.

Ralls K, Ballou JD, Templeton A (1988) Estimates of lethal equivalents and the cost of inbreeding in mammals. Conservation Biology, 2, 185-193.

Raymond M, Rousset F (1995) GENEPOP (Version 1.2): Population genetics software for exact tests and ecumenicism. Journal of Heredity, 86, 248-249.

Rojas M (1992) The species problem and conservation: what are we protecting?

Conservation Biology, 6, 170-178.

Roy MS, Geffen E, Smith D, Ostrander EA, Wayne RK (1994) Patterns of differentiation and hybridization in North American wolflike canids, revealed by analysis of microsatellite loci.

Molecular Biology and Evolution, 11, 553-570.

143 Sambrook E, Fritsch F, Maniatis T (1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbour Laboratory Press, New York.

Serena M, Soderquist TR , Morris K (1991) Western Australian wildlife management program no. 7: the chuditch (Dasyurus geoffroii). Department of Conservation and Land

Management, Wanneroo, WA.

Sinclair EA, Danks A, Wayne AF (1996) Rediscovery of Gilbert's potoroo, Potorous tridactylus, in Western Australia. Australian Mammalogy, 19, 69-72.

Slatkin M (1995) A measure of population subdivision based on microsatellite allele frequencies. Genetics, 139, 457-462.

Sokol RR, Rohlf FJ (1981) Biometry. WH Freeman and Company, New York.

Strahan R, Ed. (1998) The Mammals of Australia. New Holland Publishers, Sydney.

Swofford DL, Selander RB (1981) BIOSYS-1: a FORTRAN program for the comprehensive analysis of electrophoretic data in population genetics and systematics. Journal of Heredity,

72, 281-283.

Tautz D (1989) Hypervariability of simple sequences as a general source for polymorphic

DNA markers. Nucleic Acids Research, 17, 6463-6471.

144 Tautz D, Renz M (1984) Simple sequences are ubiquitous repetitive components of eukaryotic genomes. Nucleic Acids Research, 12, 4127-4138.

Taylor AC, Sherwin WB, Wayne RK (1994) Genetic variation of microsatellite loci in a bottlenecked species: the northern hairy-nosed wombat Lasiorhinus krefftii. Molecular

Ecology, 3, 277-290.

Thomas O (1909) Some mammals from N.E. Kimberley, Northern Australia. The Annals and Magazine of Natural History; Zoology, Botany, and Geology, 8, 149-152.

Thomas O (1926) The local races of Dasyurus hallucatus. The Annals and Magazine of

Natural History; Zoology, Botany, and Geology, 9, 543-544.

Titterington et al. (1981) Comparison of discrimination techniques applied to a complex data set of head injured patients. Journal of the Royal Society of Statistics, Ser. A. 144, 145-

175.

Vane-Wright RI, Humphries CJ , Williams PH (1991) What to protect? Systematics and the agony of choice. Biological Conservation, 55, 235-254.

Vogler AP, DeSalle R (1994) Diagnosing units of conservation management. Conservation

Biology, 8, 354-363.

145 Waples RS (1998) Evolutionarily significant units, distinct population segments, and the

Endangered Species Act: reply to Pennock and Dimmick. Conservation Biology, 12, 718-

721.

Wright S (1951) The genetical structure of populations. Annals of Eugenetics, 1, 323-354.

Zhu D, Degnan S, Moritz C (1998) Evolutionary distinctiveness and status of the endangered

Lake Eacham rainbowfish (Melanotaenia eachamensis). Conservation Biology, 12, 80-93.

Zink RM, Kale HW (1995) Conservation genetics of the extinct dusky seaside sparrow

Ammodramus maritimus nigrescens. Biological Conservation, 74, 69-71.

Acknowledgments

I thank the many people and institutions who kindly helped collect samples used in this study:

C. Belcher, S. Burnett, J. Byron, N. Cooper, F. Craven, R. Darr, R. Dickens, J. Dixon, B.

Dowling, L. Frigo, C. Gallagher, L. Gibson, J. Giles, J. Griffiths, G. Hall, M. Jones, A. Kelly,

B. Lewis, D. Longbottom, L. Leung, K. Morris, D. Moyle, M. Murray, M. Oakwood, M.

Pennay, L. Pope, S. Priestly, J. Rainbird, B. Read, J. Seebeck, J. Titmarsh, L. Vogelnest, B.

Walker, The Australian Museum, The Western Australian Museum, The Museum of Victoria, the Queen Victoria Museum, Featherdale Wildlife Park, Trowunna Wildlife Park, Perth Zoo, and Taronga Zoo. I also thank State Forests of New South Wales, the NSW National Parks

146 and Wildlife Service, the owners, managers and staff of Barrington Guest House and W.

Greville who provided assistance in many ways. Thanks also go to C. Dickman and J. Giles for providing comments that improved this paper. Funding for this work was received from an

Australian Postgraduate Research Award, the Estate of W.V. Scott, Barrington Guest House, and the Zoological Parks Board of New South Wales.

147 Chapter 5. Phylogeographical Population Structure of Tiger Quolls

Dasyurus maculatus (Dasyuridae:Marsupialia), an Endangered

Carnivorous Marsupial

Tiger quolls, Dasyurus maculatus, are the largest carnivorous marsupials still extant on the mainland of Australia, and occupy an important ecological niche as top predators and scavengers. Two allopatric subspecies are recognized, D. m. gracilis in north Queensland, and D. m. maculatus in the southeast of the mainland and Tasmania. D. m. gracilis is considered endangered while D. m. maculatus is listed as vulnerable to extinction; both subspecies are still in decline. Phylogeographic subdivision was examined to determine evolutionarily significant units (ESUs) and management units (MUs) among populations of tiger quolls to assist in the conservation of these taxa. Ninety-three tiger quolls from nine representative populations were sampled from throughout the species range. Six nuclear microsatellite loci and the mitochondrial DNA (mtDNA) control region (471 bp) were used to examine ESUs and MUs in this species. I demonstrated that Tasmanian tiger quolls are reciprocally monophyletic to those from the mainland using mtDNA analysis, but D. m. gracilis was not monophyletic with respect to mainland D. m. maculatus. Analysis of microsatellite loci also revealed significant differences between the Tasmanian and mainland tiger quolls, and between D. m. gracilis and mainland D. m. maculatus. These results indicate that Tasmanian and mainland tiger quolls form two distinct evolutionary units but that

D. m. gracilis and mainland D. m. maculatus are different MUs within the same ESU. The two marker types used in this study revealed different male and female dispersal patterns and indicate that the most appropriate units for short-term management are local populations. A

148 revised classification and management plan is needed for tiger quolls, particularly in relation to conservation of the Tasmanian and Queensland populations.

Introduction

There have been various attempts to formalise the relationship between conservation status and taxonomic status, although the area is still controversial (Vane-Wright et al. 1991; Crozier

1992; Dizon et al. 1992; Rojas 1992; Vogler and DeSalle 1994; Crozier 1997). Two widely applied concepts are the evolutionarily significant unit (ESU) and the management unit (MU).

ESUs were proposed as a means of clarifying the conservation agenda of a number of large charismatic taxa that were being conserved as separate subspecies in zoological parks, but where there was limited biological support for this subspecific status (Ryder 1986). The genetic basis of ESUs has been defined as conservation units (populations or groups of populations) showing significant differences in allele frequencies at nuclear loci and reciprocal monophyly at mitochondrial loci (Moritz 1994). ESUs, therefore, reflect historic phylogeographic separation of different populations. The goal of designating these distinct units is to allow their separate management so as to retain the long term evolutionary diversity of the species. MUs, on the other hand, are based on allele frequency differences regardless of the phylogenetic relationships of the alleles. These units are more appropriate for shorter term conservation goals such as population monitoring.

The definitions of ESUs and MUs proposed by Moritz (1994) suggest that examination of both nuclear and mitochondrial loci are pertinent to determining these conservation

149 units. Nuclear microsatellites and the mitochondrial control region are both extremely effective genetic markers for examining population level variation, differentiation, and structure (e.g.

Taylor et al. 1994; Houlden et al. 1996; Pope et al. 1996; Moritz et al. 1997; Duvernell and

Turner 1998; Zhu et al. 1998).

Tiger quolls, Dasyurus maculatus, are the largest carnivorous marsupials remaining on the mainland of Australia and are now found in several geographically isolated regions (Figure 20).

There are two currently recognized subspecies: the smaller D. m. gracilis which is restricted to north Queensland, and the larger D. m. maculatus which is found in the southeast of the mainland and in Tasmania (Figure 20). Ramsay (1888) originally described the morphology of the north Queensland taxon from a stuffed specimen and gave it specific status as D. gracilis which was later redefined as a subspecies of D. maculatus (Tate 1952). Examination of museum specimens indicates that there is a general trend for increased body size in animals from higher latitudes (Jones 1997). Since European settlement in Australia, both subspecies have suffered a decline in abundance and range; it is thought that D. m. gracilis has declined by approximately 20% and that D. m. maculatus has declined by as much as 50% in range

(Maxwell et al. 1996).

The reasons for this species’ continued decline most likely include habitat loss, direct and indirect competition with introduced predators, and continued hunting, trapping, and poisoning by humans (Mansergh 1984). In addition, tiger quoll numbers appear to be affected by the introduced poisonous cane toad, Bufo marinus, in areas where ranges of these two species overlap (Burnett 1997). Currently, both subspecies of tiger quolls

150 Figure 20. Map of Australia showing the current (black) and past (black plus gray) distribution of D. maculatus, the distribution of the subspecies, and sampling localities. Population names are as in Table 11.

Dasyurus maculatus gracilis QLD

MAN COP BAR BGH CSF Dasyurus BAD maculatus SUG maculatus TAS

are listed by the World Conservation Union (IUCN). D. m. gracilis is considered to be endangered while D. m. maculatus is listed as vulnerable to extinction (Maxwell et al. 1996).

Despite the continuing decline of tiger quolls and the apparent need for conservation action, there has been no previous molecular examination of the phylogeographical structure within this species. The aim of this study was to investigate the genetic diversity and divergence within the species in order to evaluate whether there are evolutionarily distinct units and

151 management units whose recognition would aid in sound conservation management of this taxon.

Materials and Methods

Samples and DNA Extractions

Samples for this study were collected as ear biopsies or tail tips stored in 80% ethanol, whole blood, or liver biopsies from road-killed animals stored frozen at -20° C. Total genomic

DNA was prepared by standard phenol-chloroform extraction followed by ethanol precipitation (Sambrook et al. 1989). Samples from nine populations representing both subspecies were available for this study (Figure 20; Table 11). The samples from central

Tasmania (TAS, n = 11) and the Barrington Tops area of New South Wales (BAR, n = 11) were from road-killed animals and were collected opportunistically; these samples came from a more wide-spread area than other populations. One group of individuals (MAN, n = 5) was from stock originally from the Mann River Nature Reserve of NSW but had been bred in captivity for a number of generations. In total, 93 individuals representing 9 populations and two subspecies were analysed.

Control Region Amplification, Screening and Sequencing

The 5’ end of the mtDNA control region (between the tRNApro locus and the central conserved domain) was chosen for analysis because of its wide utility as a population marker

(Pope et al. 1996; O'Corry-Crowe et al. 1997; Jerry et al. 1998). Polymerase chain reaction (PCR) amplification of the 5’ portion of the control region was achieved using primers mt15999L and mt16498H which were designed to amplify DNA from a

152 ° 15’ ° 59’ ° 09’ ° 08’ ° 44’ ° 07’ ° 57’ Long (S) 16 31 32 32 29 36 36 ° 02’ ° 48’ ° 42’ ° 31’ ° 04’ ° 33’ ° 22’ Map Reference ______Lat (E) 145 151 151 151 wide area 152 149 148 wide area 2 2 ) n 9 22 17 5 6 3 microsatellite 1 Sample size ( ______mtDNA 119 21 16 12 105 5 3 11 11 11 Population QLD COP BGH CSF BAR MAN BAD SUG TAS 4 3 3 n =5,CSF) were excluded from microsatellite analyses. Location 1. Mt Windsor, Qld 2. Copeland, NSW 3. Barrington Guest House, NSW 4. Chichester State Forest, NSW 5. Barrington Tops area, NSW 6. Mann River Nature Reserve, NSW 7. Badja State Forest, NSW 8. Suggan Buggan, Vic 9. Central Tasmania n =6, BGH) ( Subspecies, sample populations, locations, and sizes used in this study. Pouch young ( Represents road-killed animals collected opportunistically and covers more widely dispersed areas than other populations. Captive bred stock. Two additional individuals were analysed at the mtDNA control region, but not part of any population. Table 11. Subspecies D. m. gracilis D. m. maculatus 1 2 3 4

153 wide range of marsupials (Fumagalli et al. 1997). DNA (~20-100 ng) was amplified in a 25

mL reaction containing 2.5 mM MgCl2, 0.75 units of Tth Plus DNA polymerase (Biotech

International, Ltd.), 67 mM Tris-HCl pH 8.8, 16.6 mM [NH4]2SO4, 0.45% Triton X-100,

0.2 mg/mL gelatine, 0.1 mM each primer, and 0.2 mM each dNTP. Thermal cycling was performed by denaturing at 94 °C for 2 min followed by 35 cycles of denaturation (94 °C, 45 sec) annealing (50 °C, 45 sec) and extension (72 °C, 1 min) with a final extension of 5 min at

72 °C.

Mitochondrial DNA amplification products were screened for variants using heteroduplexing in combination with temperature gradient gel electrophoresis (TGGE, Campbell et al. 1995;

Elphinstone and Baverstock 1997; Heslewood et al. 1998). Temperature gradient gels were not directly comparable, therefore each haplotype detected on each gel was sequenced.

Mitochondrial haplotypes were sequenced from PCR products using either the ABI

PRISMTM Dye Terminator or the BigDyeTM Terminator Cycle Sequencing Ready Reaction kits with AmpliTaqâ DNA polymerase FS and primers mt15999L and mt16498H.

Sequencing was performed on an ABI 377 automated sequencer. All sequences have been deposited in GenBank (accession numbers AF082751-AF082770).

Microsatellite Amplification and Screening

Six highly polymorphic (CA)n microsatellite loci specifically designed for quolls were identified and described (Q1.3, Q3.1.2, Q3.3.1, Q3.3.2, Q4.4.2, and Q4.4.10; Firestone, 1999).

Microsatellite alleles were identified by PCR amplification using a radioactively labelled [g-

33P]ATP primer as described (Firestone, 1999). Products were separated on 6%

154 polyacrylamide gels containing 50% w/v urea; 4 mL of PCR product was mixed with an equal volume of formamide loading dye and heated to 80 °C for 3 min, and then loaded on a sequencing gel. M13 control DNA was sequenced (USB Sequenase 2.0 kit) and electrophoresed as a size marker on all gels to allow exact scoring of allele sizes. Gels were fixed in 1:1 glacial acetic acid:100% methanol, dried, and exposed to Kodak X-OMAT AR film for 24-72 h.

Statistical Analyses

MtDNA sequences were aligned using the comparative alignment algorithm in Sequence

Navigator (Perkin Elmer, Applied Biosystems) and adjusted manually to eliminate excess gaps. Haplotypic diversity, nucleotide diversity and nucleotide divergence within and among populations was calculated using REAP 4.0 (McElroy et al. 1992) while the Monte Carlo algorithm in REAP was used to test for geographic heterogeneity in haplotypic frequency distributions using 5000 replicates. In addition, the partitioning of sequence variation between subspecies and among different geographic groupings of populations was assessed using

AMOVA ver. 1.55 (Excoffier et al. 1992). This program provides an analysis of haplotypic diversity at different hierarchical levels and produces F-statistics analogous to Wright’s

(1951) F-statistics. The significance of these statistics were tested using 1000 randomized permutations which requires few assumptions about the structure of the data.

Phylogenetic relationships among the mtDNA haplotypes were assessed by various means.

Firstly, Tamura (1992) distances generated from the sequence data were used to build a neighbour-joining tree using MEGA (Kumar et al. 1993). The significance of

155 this branching pattern was assessed by 500 bootstrap replicates. Secondly, inference of evolutionary relationships was made using maximum parsimony analysis in PAUP 3.1

(Swofford 1993). All minimum trees were saved using the heuristic search algorithm and a bootstrapped consensus tree (2000 replicates) was generated. Finally, the relationships of haplotypes was examined also by means of a minimum spanning network analysis (Excoffier and Smouse 1994) which allows haplotypes to be connected by a series of mutational events.

Microsatellite genetic diversity was measured as the number of alleles (A), the percentage of

loci that were polymorphic (%P), and the observed (Ho) and unbiased expected (HE) heterozygosities using BIOSYS ver. 1.7 (Swofford and Selander 1981). Heterogeneity of microsatellite allele frequencies was tested among all population pairs using an unbiased estimate of the exact Fisher test (GENEPOP ver. 3.1b; Raymond and Rousset 1995) with significance adjusted for multiple pairwise comparisons by a sequential Bonferroni correction

(Rice 1989). The distribution of allele frequency variance among populations was analysed using both allele frequencies (q, as implemented in FSTAT ver. 1.2; Goudet 1995, Weir and

Cockerham 1984) and allele frequencies in conjunction with allele size information (r statistics in RST CALC ver. 2.2; Goodman 1997, Slatkin 1995). The significance of both q and r statistics was tested by 1000 permutations and 1000 bootstrap replicates.

156 Results

Control Region Variation

Approximately 471 base pairs of sequence from the 5’ end of the mtDNA control region were obtained from 20 Dasyurus maculatus individuals. Homologous control region sequence was also obtained from one other species of quoll and used as an outgroup (eastern quoll, D. viverrinus). The outgroup species had a longer sequence than that of tiger quolls which necessitated the insertion of gaps for alignment (final aligned sequence length ~507 bp).

Twelve different control region haplotypes were identified by heteroduplex/TGGE and sequencing in tiger quolls; 36 variable sites were found (7.6%), 19 of which were phylogenetically informative (Figure 21). Transitions accounted for 80.2% of the total nucleotide substitutions and all sequences had a high adenine:thymine ratio (>74%) in keeping with other known mammalian mtDNA control region sequence data (Wolstenholme 1992;

Janke et al. 1994; Gemmell et al. 1996).

The average haplotypic diversity within populations (HD, Table 12) was 0.3860 ± 0.0081

(range: 0.0000 to 0.7455) while the average within population nucleotide diversity (P, Table

12) was 0.0060 ± 0.0001 (range: 0.0000 to 0.0124). Average nucleotide divergence and nucleotide diversity between populations were 0.0139 (range: -0.0009 to 0.0398) and 0.0199

(range: 0.0061 to 0.0431) respectively.

157 Figure 21. Sequence variation among 12 mtDNA control region haplotypes detected among 93 individual tiger quolls. Only variable positions are shown, as indicated by numbers at the top. Complete sequences are available from NCBI GenBank (accession numbers AF082751-AF082770). Number of individuals with each haplotype is listed in parentheses. A dot (.) indicates identity with the top sequence; letters indicate nucleotide substitutions.

1111111112222222233333344444 112335690011122691222266614566800345 469456872927924525167945800201239821 Consensus TTCTATACCCGTGTCTTTAACAATTGAGCATTTTTC Hap A (19) ...... C..T...... T...... T Hap B (35) ...... C...... C...... C. Hap C (7) ...... T...A..C..GGT...CT...... Hap O (5) ??.A...... G..C...... C.. Hap Q (1) ...... C...... Hap J (3) ...... C...... GCC...T...C... Hap G (9) C...... T...A.....G.T...C...... Hap H (3) ...... C...... C...... Hap E (3) ..T.GAT.....AC.....GT...... A.TC..... Hap AC(2) ..T.GATT.T..AC.....GT...... A.T...... Hap D (5) .CT.GA....A.AC.....GT...... A.T.C.... Hap F (1) ..T.GATT....AC.....GT...... A.T......

158 4

P 0.005 0.009 0.004 0.007 0.012 0.011 0.000 0.000 0.006 3 HD Tot. 11 0.327 91 9 0.556 21 0.257 16 0.500 10 0.689 5 0.400 5 0.000 3 0.000 11 0.745 AC - 2 O 5 -- J --- 3 5 H 2 G 9 2 ---- Haplotypes F E ------5 3 1 CD ------4 7 5 3 1 9 2 3 5 2 18 6 - ______AB ------23 610 31 1 5------2 19 35 1 Sample Locations 1. QLD 2. COP 3. BGH 4. CSF 5. BAR 6. MAN 7. BAD 8. SUG 9. TAS = Nucleotide diversity within populations. Sample locations follow that of Table 11. Haplotype Q was not included in this analysis since it found only an isolated road-killed individual and did represent a P Haplotype H was also found in one individual from Victoria, but not SUG; this included population HD = Haplotypic diversity within populations. Table 12. Haplotype frequencies, haplotype diversity, and nucleotide diversity in nine tiger quoll populations. Subspecies D. m. gracilis D. m. maculatus Totals 1 2 population. 3 4 5 analyses

159 Geographic Distribution of Control Region Haplotypes

In general, mtDNA haplotypes had limited distributions which were often unique to a single population (Table 12). Populations usually possessed from one to three different haplotypes, though the TAS population possessed four haplotypes. Haplotypes ‘A,’ ‘B,’ and ‘C’ were found in more than one population (COP, BGH, CSF, BAR, MAN), however, the majority of these populations were in relatively close proximity to one another, and all were within NSW.

Haplotype ‘H’ was the only one found in two widely geographically separate areas

(Queensland and Victoria).

Monte Carlo simulation indicated that there was a highly significant partitioning of haplotypes among all populations (c2 = 423.49, 80 d.f., P £ 0.0001) as well as significant partitioning of haplotypes among the four Barrington Tops populations (c2 = 17.19, 6 d.f., P £ 0.008).

Significant differences in the Barrington region primarily were due to the different frequencies of haplotypes ‘A’ and ‘B’ in the BGH and CSF populations ( c2 = 9.26, 1 d.f., P = 0.001).

Phylogeographic Relationships of mtDNA Haplotypes

Neighbour joining analysis showed that Tasmanian and mainland tiger quolls were reciprocally monophyletic with respect to each other. This was strongly supported by 98% of bootstrap replicates (Figure 22). North Queensland D. m. gracilis haplotypes, however, were polyphyletic with respect to southern mainland D. m. maculatus haplotypes and fell within the mainland clade. Results from parsimony analysis largely followed the same pattern although some of the internal nodes were collapsed (data not shown).

160 ) D. maculatus F HAP A HAP Q HAP O HAP H HAP B HAP J HAP C HAP G HAP D HAP E HAP AC HAP HAP AV VIC NSW (Dv) as an outgroup. All Dm ( NSW NSW TAS TAS NSW TAS TAS QLD, VIC NSW D. viverrinus QLD 68ÀÄÄÄÄÄÄÄÄÄÄ

is approximately equal to the distance of 0.01 ÄÄÄÄÄÄÄÄÄÄÄ 94Ú ÚÄÄÄÄÄ´ ÚÄ´ ÀÄÄ ³ 68³ ÀÄÄÄÄÄÄÄÄÄ ÚÄÄÄÄÄÄ´ 66³ ³ÚÄÄÄ ÚÄÄÄÄÄ´ ³³ ³ À´ ÚÄÄ ³ ÀÄ´ Dm ³ ÚÄÄÄ\\ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ´ ³ ³ ÚÄÄÄÄÄÄ ³ ÀÄÄÄ´ ³ 71ÀÄÄÄ ³ ³ ÚÄÄÄÄÄÄÄ ³ ³ ÚÄÄ ³ ÀÄÄÄÄÄÄÄÄÄÄÄÄÄ´ ³ 98ÀÄ´ ÚÄÄ ³ 92ÀÄ´ ³ 65À ³ ÀÄÄÄ\\ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ Dv Scale: Figure 22. Phylogenetic relationships among the mtDNA haplotypes. Neighbour-joining tree based on Tamura distances. Significance of branching pattern is indicated by the proportion of bootstrap replicates in which clade to right appears; only values over 60% are shown. Haplotypes are as in Figure 21. This tree was rooted with the eastern quoll, haplotypes are also listed by States in which the haplotype occurs.

161 A minimum spanning network is an alternative way of looking at phylogenetic relationships by illustrating the number of nucleotide substitutions among haplotypes in an interconnecting form rather than a simple bifurcating tree (Figure 23). This analysis produced a relatively uncomplicated network in which the only alternative connections were either between haplotypes ‘B’ and ‘J’ or between haplotypes ‘H’ and ‘J.’ On average, within-mainland and within-Tasmanian haplotypes differed from one another by 4.0 and 2.3 nucleotide substitutions, respectively. However, there were a minimum of 11 nucleotide changes between the Tasmanian and mainland haplotypes which represents a three- to five-fold increase in base pair substitutions. Similarly to the distance-based neighbour joining tree, D. m. gracilis haplotypes did not branch off from other mainland D. m. maculatus haplotypes.

Control Region Population Differentiation

Table 13 shows the results of various hierarchical and non-hierarchical analyses of variance based on mtDNA haplotype diversity. When no a priori hierarchy was set up, 64% of the total variance present was due to that among populations. There was a highly significant

degree of genetic partitioning among these populations (FST = 0.644; P < 0.001) indicating strong barriers to gene flow and a high degree of genetic structure. When the four Barrington

Tops populations (COP, BGH, CSF, BAR) were analysed separately, 84% of the variance was found within individual populations. However, there was also significant subdivision

among the Barrington Tops populations (FST = 0.159; P < 0.001) indicating that these populations could not be pooled for further analysis, although they were not highly differentiated. In a small locality such as the Barrington region, drift or migration may be more important than the accumulation of

162 Figure 23. Minimum spanning network of 12 different control region haplotypes found in tiger quolls. Cross hatches represent nucleotide substitutions between haplotypes. Haplotypes from the mainland are indicated in open circles, those from Tasmania are indicated by grey. D. m. gracilis haplotypes (G and H) are found in the same cluster as the southern mainland D. m. maculatus haplotypes.

J B H A O Q G C

D

F E AC

163 mutations in defining population units. Thus, I also examined subdivision among the Barrington

Tops populations excluding mutational information. In this analysis, populations were still

significantly subdivided, although again, not highly differentiated (FST = 0.164, P £ 0.001).

When populations were divided into geographical groupings (i.e. Tasmania vs. all mainland populations, Table 13), 81% of the total haplotypic diversity was found among populations or

groups which again indicates highly structured populations (FST = 0.806; P < 0.001); 61% of the total haplotypic diversity was due to variation distributed between the geographical

groupings (FCT = 0.606; P < 0.001) and 20% of the total diversity was explained by variation among populations within regions (FSC = 0.509; P < 0.001).

Different conclusions were drawn when populations were divided by subspecific classification

(i.e. D. m. maculatus from Tasmania and the mainland versus D. m. gracilis, Table 13).

Only 3% of the variation present was due to that between subspecific groupings (FCT =

0.027; P > 0.05) indicating that differentiation between subspecies was very low relative to

either differentiation among populations within each subspecies (FSC = 0.642; P < 0.001) or differentiation within populations (FST = 0.651; P < 0.001). Results from this analysis support the phylogeographical findings in that the currently designated subspecies are poorly defined units which do not reflect the actual genetic divergence within the species.

164 P >0.05 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 64.42 35.58 15.87 84.13 2.69 62.45 34.86 60.55 20.07 19.37 52.15 8.11 39.73 % Total 0.0059 0.0032 0.0007 0.0036 0.0003 0.0058 0.0033 0.0102 0.0034 0.0033 0.0044 0.0007 0.0034 Variance Estimated Component Mean 0.061 0.003 0.013 0.004 0.074 0.059 0.003 0.238 0.035 0.003 0.093 0.013 0.003 Squares 0.4862 0.2670 0.0381 0.1868 0.0739 0.4124 0.2670 0.2381 0.2481 0.2670 0.0933 0.0381 0.2102 Sums of Squares 8 3 1 7 1 7 1 3 df 82 52 82 82 62 1 AP AP AP AP AP AG AG AG WP WP WP WP WP Source of Variation v. Groups D. m. gracilis Tasmania v. Queensland v. Barrington none none D. m. maculatus Mainland Division Species Barrington Subspecies Geographical AG = among groups, AP populations, WP within populations. Table 13. Nested analysis of molecular variance for mtDNA in tiger quolls. 1

165 Hierarchical comparisons among mainland tiger quolls (i.e. excluding TAS) showed significant

levels of subdivision (FST = 0.603; P < 0.001) with 52% of the total haplotypic variance existing among groups (FCT = 0.522; P < 0.001). Differences among populations within groups were also significant (FSC = 0.170; P < 0.001), although low relative to that among groups and within populations. Results from this analysis indicate that there are significant differences in haplotype frequencies between QLD and Barrington Tops populations (Table

13).

Microsatellite Variation

Six microsatellite loci were evaluated for genetic diversity as measured by the number of alleles, percent of polymorphic loci, and observed and expected heterozygosities (Table 14).

All loci were polymorphic although one population (MAN) was monomorphic at locus Q1.3.

The overall average number of alleles per locus calculated for all populations was relatively low for microsatellites (3.5 alleles). However some individual populations had six to eight alleles per locus and most loci had seven or more total alleles (see Appendix 6). In general, the populations with the highest number of individuals sampled (TAS, BGH, CSF, BAR) had the greatest mean number of alleles/locus. However the QLD population had lower levels of allelic diversity in comparison with these other populations, which may indicate some loss of genetic variability. Some populations (MAN, BAD, SUG) had very low numbers of alleles but this is possibly due, at least in part, to small sample sizes.

166 Table 14. Microsatellite diversity in tiger quolls averaged over six loci. Means and standard errors, in parentheses, are given.

1 2 3 4 5 6 Population n A % P HO HE

1. QLD 11.7 (0.2) 2.50 (0.22) 100 0.43 (0.07) 0.49 (0.05)

2. COP 8.5 (0.3) 4.33 (0.49) 100 0.62 (0.12) 0.66 (0.08)

3. BGH 15.7 (0.2) 4.67 (0.49) 100 0.66 (0.05) 0.67 (0.04)

4. CSF 12.0 (0.0) 4.67 (0.56) 100 0.65 (0.06) 0.67 (0.04)

5. BAR 9.3 (0.9) 4.67 (0.87) 100 0.59 (0.07) 0.62 (0.08)

6. MAN 4.5 (0.2) 2.67 (0.42) 83.3 0.43 (0.10) 0.50 (0.10)

7. BAD 5.8 (0.2) 2.50 (0.34) 100 0.37 (0.12) 0.47 (0.07)

8. SUG 3.0 (0.0) 3.67 (0.22) 100 0.61 (0.10) 0.58 (0.09)

9. TAS 10.8 (0.2) 3.83 (0.60) 100 0.51 (0.10) 0.59 (0.09)

1 Population codes follow Table 11 2 Sample size per locus 3 Number of alleles per locus 4 Percent of loci polymorphic 5 Observed heterozygosity by direct count 6 Unbiased expected heterozygosity

Microsatellite Population Differentiation

Tests of heterogeneity in allele frequencies revealed that there were significant differences in at least one locus between most pairs of populations after sequential Bonferroni correction (P £

0.0016; Table 15). Populations in the Barrington Tops area, however, were not significantly differentiated from each other based on allele frequency distributions with the exception of one pairwise comparison at one locus (between BAR and BGH at locus Q4.4.10), which may be due to chance alone. The QLD and TAS

167 populations had a greater number of significant pairwise differences in allele frequencies than all other populations (Table 15).

Table 15. Pairwise measures of microsatellite differentiation in nine populations of tiger quolls. Below the diagonal: the number of loci (out of six) with significant differences in allele frequency distributions (unbiased estimate of the exact Fischer test). Above the diagonal: pairwise q values (significant values in bold). All pairwise comparisons were adjusted by a sequential Bonferroni correction.

QLD COP BGH CSF BAR MAN BAD SUG TAS QLD - 0.156 0.221 0.221 0.273 0.445 0.398 0.429 0.298 COP 3 - 0.047 0.011 -0.003 0.271 0.167 0.184 0.117 BGH 6 0 - 0.029 0.055 0.236 0.143 0.086 0.154 CSF 4 0 0 - 0.015 0.225 0.117 0.136 0.152 BAR 4 0 1 0 - 0.252 0.123 0.112 0.123 MAN 5 1 1 1 1 - 0.360 0.267 0.338 BAD 5 1 2 1 1 2 - 0.117 0.255 SUG 5 1 0 0 0 1 0 - 0.222 TAS 5 1 3 4 1 3 3 0 -

Analysis of population subdivision was performed using both q and r since drift or migration might be more important than mutation in defining separate MUs. The proportion of variation distributed among all nine populations was low but highly significant in both cases (q = 0.171,

P < 0.001; r = 0.121, P < 0.0001). Pairwise estimates of q ranged from -0.003 to 0.445 and most values were significantly greater than zero (Table 15). However, examination of the

Barrington Tops populations (COP, CSF, BGH, BAR) showed that q was not significantly different from zero after sequential Bonferroni correction in any pairwise comparisons, indicating that there was no population subdivision among these groups (P ³ 0.004; Table 16).

These populations were then pooled and compared to TAS and QLD which showed highly significant differentiation for all pairwise comparisons (P < 0.001; Table 16).

168 Overall r values (averaged over variance components) for the four Barrington Tops populations were not significantly different from zero (r = 0.02; P = 0.13; Table 16), again indicating that there was no subdivision among these populations based on allele size/allele frequency information. Therefore, the Barrington Tops populations were pooled and compared with the TAS and QLD populations, as above. This analysis also showed a highly significant level of population structuring (r = 0.06; P = 0.001). When pairwise population comparisons were made, the TAS population was consistently significantly differentiated from both the QLD and combined Barrington Tops populations (Table 16). In this analysis, however, tiger quolls from north Queensland (QLD) were not differentiated from southern mainland tiger quolls (pooled BT; Table 16).

Table 16. Pairwise population subdivision of nuclear loci. ______Population Comparison r P q P ______Over All BT populations 0.020 0.130 0.028 >0.004

TAS x pooled BT 1 0.084 0.008 0.191 <0.001

TAS x QLD 0.131 0.003 0.298 <0.001

QLD x pooled BT 0.022 0.109 0.122 <0.001 ______1 Pooled BT = pooled Barrington Tops population (i.e. COP, BGH, CSF and BAR populations)

169 Discussion

This was the first study to examine genetic variation and differentiation within any of the large carnivorous marsupials. These analyses examined genetic partitioning and historical phylogeographic patterns in tiger quolls to evaluate MUs and ESUs within this species to aid in their conservation. As has been shown in a wide variety of other species (García-Moreno et al. 1996; Luikart et al. 1997; Moritz et al. 1997; Brunner et al. 1998; Pichler et al. 1998;

Walker et al. 1998; Zhu et al. 1998), both microsatellites and the mtDNA control region were highly variable and informative population markers in tiger quolls.

Within-Population Levels of Genetic Diversity

Tiger quolls exhibited moderate levels of diversity as compared to other species. Within populations of tiger quolls, the average level of nucleotide diversity (0.6%) fell within the range of that found in other marsupials from a low of 0.16% in koalas, to a high 1.36% in bilbies,

(Moritz et al. 1997; Houlden et al. 1999). Similarly, average within-population haplotypic diversity of tiger quolls (HD = 0.39) was approximately twice that of koalas (HD = 0.18), but still less than that found within bilby populations (HD = 0.95) (Moritz et al. 1997; Houlden et al. 1999). Average heterozygosity at microsatellite loci was 0.58, within the range found in koalas. However allelic diversity within tiger quoll populations (A = 3.5) was very low in comparison to that of outbred koala populations (A = 11.5) which may be partially attributed to the small sample sizes in quolls.

170 Among-Population Levels of Genetic Differentiation

Tiger quoll populations were highly differentiated from one another. Examination of the mtDNA control region showed that even those populations that were in very close geographic proximity to one another were highly subdivided. This subdivision was manifested by both

significant haplotype frequency differences and significant FST seen among the Barrington

Tops populations, and is possibly due to strong female philomatry (defined here as juvenile females remaining within their mothers’ ranges upon maturity) and the lower effective size of a maternally inherited marker (Birky et al. 1989).

Analysis of microsatellite loci presented another picture, however: populations that were in close proximity to one another (within ~50 km radius) were not significantly differentiated from each other in either the distribution of allele frequencies (1/36 pairwise comparisons) or in pairwise measures of population subdivision (q or r). At larger geographical scales, there were significant pairwise differences in the distribution of allele frequencies at most loci, and most population pairs had q values that were significantly different from zero (Table 15), indicating that these populations should be considered as separate MUs. However, even populations that were separated by large-scale geographic distances (QLD v. pooled BT)

were not detectably subdivided when allele size was considered (rST = 0.022, P = 0.109).

This lack of differentiation, may be due to the high mutation rate of microsatellites. Only populations which had clear limits to dispersal (TAS v. mainland) showed overall significant subdivision at nuclear loci using both q and r. Significant differences between Tasmanian and mainland

171 populations may be due to both the long term separation of these populations (late

Pleistocene, 12-15,000 years ago) and genetic drift.

The two marker types used in this study revealed different male and female dispersal patterns: strong female philomatry leading to localized mitochondrial lineages and male biased dispersal leading to undifferentiated populations based on nuclear loci. Female philomatry and different effective sizes of the markers may be the most likely explanation for the observed structuring, although other factors such as different mutation rates of the markers, selection, or biogeographic history may also play a role. In this case, however, mitochondrial differentiation

was nearly six times that of nuclear differentiation (fST = 0.164; qST = 0.028) indicating a strong male-biased dispersal. In addition, conclusions drawn from our analysis of different molecular markers is supported by field data. Preliminary results of radiotracking and mark- recapture studies of tiger quolls in the Barrington Tops region have shown that young female tiger quolls remain within their mothers’ home ranges, while male offspring will disperse upon maturity (Firestone, in preparation). Home range sizes of females in this area were in the order of 2 - 4 km2, which could give rise to localized female lineages of mtDNA haplotypes within valleys (Firestone, in preparation). Even at distances as short as 20 km, significantly different haplotype frequencies were evident. This pattern of sex-biased genetic structuring is also found in other species, notably sperm whales (Lyrholm et al. 1999) and ghost bats

(Worthington Wilmer et al. in press) where the high energetic requirements of breeding females is proposed as the mechanism causing the observed structure.

172 Evolutionary Distinctions and Subspeciation Among Tiger Quolls

Reciprocal monophyly at the mtDNA control region and significant differences in both nuclear and mitochondrial allele frequencies indicate long-term phylogeographic separation between

Tasmanian and mainland tiger quolls. The use of both nuclear and mitochondrial loci has shown consistent differentiation between mainland and island forms, and that these groups form two distinct ESUs. However, the currently designated subspecies, D. m. gracilis and D. m. maculatus, do not reflect the actual genetic subdivisions present within this species. D. m. gracilis and D. m. maculatus are not reciprocally monophyletic at mtDNA loci, although there are differences between these populations in the distribution of nuclear and mitochondrial allele frequencies. Thus D. m. gracilis and southern mainland D. m. maculatus are different

MUs within the same ESU.

The concept of subspecies is related to that of ESUs. Generally, subspecies are considered to be regional variants of the species. The current classification among tiger quolls distinguishes only two subspecies. The lack of taxonomic clarity has often been disastrous for the conservation of endangered taxa, as was the case for both tuataras and dusky seaside sparrows (Daugherty et al. 1990; Zink and Kale 1995). The existing taxonomic status of tiger quolls in Tasmania should be reassessed and elevated to the subspecific level to more accurately reflect the significant divergence between the two ESUs. On the other hand, our results call into question the importance of the current division between D. m. gracilis and D. m. maculatus. The morphological differences between these two subspecies probably reflect a clinal adaptation to climatic differences and a manifestation of Bergmann’s Rule (where body size increases at higher latitudes).

173 Conservation Status, Management Issues, and Recommendations

MUs are short-term conservation units. In the case of localized population decline or extinction, these populations would be unlikely to recover in the absence of active management

(i.e. translocations) (Moritz 1994, Moritz et al. 1995). Most of the populations that were examined are separate MUs based on pairwise comparisons of nuclear allele frequencies.

Continued monitoring of these populations would enable conservation managers to take appropriate action in the case of population decline.

ESUs, on the other hand, recognize the historic separation and divergence between groups; therefore their separate management allows for continued divergence and evolution of the species. The possibility of relocations between ESUs is very real in tiger quolls: in the early

1940s tiger quolls from Tasmania and the mainland were actively bred and their offspring were released onto the mainland (Fleay 1940; Fleay 1948). In addition, tiger quolls are currently held and bred in captivity in a number of Australian institutions and there have been proposals to reintroduce tiger quolls into areas of their former range. The recognition of distinct evolutionary units will allow managers and agencies to make better informed management decisions in the future and avoid commingling of individuals from genetically divergent ESUs.

Although the Tasmanian tiger quoll population numbers appear to be stable at present, the conservation status of this population may be less secure than previously recognized due to our suggested taxonomic ‘splitting.’ Furthermore, it is possible that the conservation status of the

Queensland population may be regarded as more secure now due to taxonomic ‘lumping.’

However, designation of ESUs should not be viewed as

174 the only means of assessing the conservation status of species, but rather as complementary to other methods of assessment (Moritz et al. 1995). Thus the listing of the north Queensland population as endangered should stand, because although it is not a separate ESU, that population is a distinct MU based on significant differences in frequencies of both mtDNA haplotypes and nuclear alleles.

Three main recommendations emerge from our findings. First, Tasmanian populations of tiger quoll should be recognized as distinct conservation units separate from those on the mainland.

The taxonomy should reflect this and the Tasmanian populations should be elevated to the subspecies level. Second, the status of north Queensland tiger quoll populations should stand since populations, not species, are the units of conservation. Queensland tiger quolls are a separate MU based on mitochondrial and nuclear allele frequency differences; they are now isolated with extensive barriers to gene flow, and possibly have different adaptations as evidenced by their smaller body size. Lastly, nuclear and mtDNA markers provided contrasting estimates of male and female dispersal and population subdivision among localized populations. If we were to examine nuclear allele frequencies only, we would conclude that some pairs of populations were not distinct, that there was a high level of gene flow between them, and thus they would not be considered separate units for management. However, examination of mtDNA haplotype frequencies would tell us that populations even in very close proximity were separate units for management. Therefore, if females in the populations were to decline (due to demographic stochasticity or other factors), it is unlikely that a population would be recolonized except over extremely short distances. Thus the most appropriate management unit for this species is the localized population.

175 It has been suggested by some conservation managers that genetic analyses can have little value for management of this species, in particular, since managers would probably conserve such widely disjunct populations separately anyway. However, separate management has not always been the case (Fleay 1940; Fleay 1948). In addition, these findings have shown that the current subspecies do not indicate the true levels of genetic partitioning within the species.

This study has also shown that it is the combination of genetic, geographic, and phenotypic information that provides the best definition of conservation units in tiger quolls. Defining MUs and ESUs among tiger quolls should assist managers in conserving this species considerably.

References

Birky CW, Fuerst P, Maruyama T (1989) Organelle gene diversity under migration, mutation, and drift: equilibrium expectations, approach to equilibrium, effects of heteroplasmic cells, and comparison to nuclear genes. Genetics, 121, 613-627.

Brunner PC, Douglas MR, Bernatchez L (1998) Microsatellite and mitochondrial DNA assessment of population structure and stocking effects in Arctic charr Salvelinus alpinus

(Teleostei: Salmonidae) from central Alpine lakes. Molecular Ecology, 7, 209-223.

176 Burnett S (1997) Colonizing cane toads cause population declines in native predators: reliable anecdotal information and management implications. Pacific Conservation Biology, 3, 65-

72.

Campbell NJH, Harriss FC, Elphinstone MS, Baverstock PR (1995) Outgroup heteroduplex analysis using temperature gradient gel electrophoresis: high resolution, large scale, screening of DNA variation in the mitochondrial control region. Molecular Ecology, 4, 407-418.

Crozier RH (1992) Genetic diversity and the agony of choice. Biological Conservation, 61,

11-15.

Crozier RH (1997) Preserving the information content of species: genetic diversity, phylogeny, and conservation worth. Annual Review of Ecology and Systematics, 28, 243-268.

Daugherty CH, Cree A, Hay JM, Thompson MB (1990) Neglected taxonomy and continuing extinctions of tuatara (Sphenodon). Nature, 347, 177-179.

Dizon AE, Lockyer C, Perrin WF, DeMaster DP, Sisson J (1992) Rethinking the stock concept: a phylogeographic approach. Conservation Biology, 6, 24-36.

177 Duvernell DD, Turner BJ (1998) Evolutionary genetics of Death Valley pupfish populations: mitochondrial DNA sequence variation and population structure. Molecular Ecology, 7, 279-

288.

Elphinstone MS, Baverstock PR (1997) Detecting mitochondrial genotypes by temperature gradient gel electrophoresis and heteroduplex analysis. BioTechniques, 23, 982-986.

Excoffier L, Smouse PE (1994) MINSPNET. Available from: http://anthropologie.unige. ch/LGB/software/win/min-span-net/

Excoffier L, Smouse PE, Quattro JM (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics, 131, 479-491.

Firestone K (1999) Isolation and characterization of microsatellites from carnivorous marsupials (Dasyuridae:Marsupialia). Molecular Ecology, 8, 1084-1086.

Fleay D (1940) Breeding of the tiger-cat. Victorian Naturalist, 56, 159-163.

Fleay D (1948) Australia's marsupial tiger cat: a creature of increasing rarity and extreme pugnacity. Wildlife Australian Nature Magazine, 10, 459-464.

178 Fumagalli L, Pope LC, Taberlet P, Moritz C (1997) Versatile primers for the amplification of the mitochondrial DNA control region in marsupials. Molecular Ecology, 6, 1199-1201.

García-Moreno J, Matocq MD, Roy MS, Geffen E, Wayne RK (1996) Relationships and genetic purity of the endangered Mexican wolf based on analysis of microsatellite loci.

Conservation Biology, 10, 376-389.

Gemmell NJ, Western PS, Watson JM, Marshall Graves JA (1996) Evolution of the mammalian mitochondrial control region-comparisons of control region sequences between monotreme and therian mammals. Molecular Biology and Evolution, 13, 798-808.

Goodman SJ (1997) Rst Calc: a collection of computer programs for calculating estimates of genetic differentiation from microsatellite data and determining their significance. Molecular

Ecology, 6, 881-885.

Goudet, J (1995) F-STAT: A computer program to calculate F-statistics. Journal of

Heredity, 86, 485-486.

Heslewood MM, Elphinstone MS, Tidemann S, Baverstock PR (1998) Myoglobin intron variation in the Gouldian finch Erythrura gouldiae assessed by temperature gradient gel electrophoresis. Electrophoresis, 19, 142-151.

179 Houlden BA, Costello BH, Sharkey D, Fowler EV, Sherwin WB, Melzer A, Ellis W, Carrick

F, Baverstock PR, Elphinstone MS (1999) Phylogeographic differentiation in the mitochondrial control region in the koala, Phascolarctos cinereus (Goldfuss, 1817) reveals a species comprising a single evolutionarily significant unit. Molecular Ecology, 8, 999-1011.

Houlden BA, England PR, Taylor AC, Greville WD, Sherwin WB (1996) Low genetic variability of the koala Phascolarctos cinereus in south-eastern Australia following a severe population bottleneck. Molecular Ecology, 5, 269-281.

Janke A, Feldmaier-Fuchs G, Thomas WK, von Haeseler A, Pääbo S (1994) The marsupial mitochondrial genome and the evolution of placental mammals. Genetics, 137, 243-256.

Jerry DR, Dow TA, Elphinstone MS, Baverstock PR (1998) Historical and contemporary maternal population structuring in the endangered Hastings River mouse (Pseudomys oralis).

Conservation Biology, 12, 1017-1022.

Jones M (1997) Character displacement in Australian dasyurid carnivores: size relationships and prey size patterns. Ecology, 78, 2569-2587.

Kumar S, Tamura K , Nei M (1993) MEGA: Molecular Evolutionary Genetics Analysis.

The Pennsylvania State University, University Park, PA, 16802.

180 Luikart G, Painter J, Crozier RH, Westerman M, Sherwin B (1997) Characterization of microsatellite loci in the endangered long-footed potoroo Potorous longipes. Molecular

Ecology, 6, 497-498.

Lyrholm T, Leimar O, Johanneson B, Gyllensten U (1999) Sex-biased dispersal in sperm whales: contrasting mitochondrial and nuclear genetic structure of global populations.

Proceedings of the Royal Society of London, B, 266, 347-354.

Mansergh I (1984) The status, distribution and abundance of Dasyurus maculatus (tiger quoll) in Australia, with particular reference to Victoria. The Australian Zoologist, 21, 109-

122.

Maxwell S, Burbidge AA, Morris K, Eds. (1996) The 1996 action plan for Australian marsupials and monotremes. Wildlife Australia, Canberra.

McElroy D, Moran P, Birmingham E, Kornfield I (1992) REAP: the restriction enzyme analysis package. Journal of Heredity, 83, 157-158.

Moritz C (1994) Defining 'Evolutionarily Significant Units' for conservation. Trends in

Ecology and Evolution, 9, 373-375.

Moritz C, Heideman A, Geffen E, McRae P (1997) Genetic population structure of the greater bilby, Macrotis lagotis, a marsupial in decline. Molecular Ecology, 6, 925-936.

181 Moritz C, Lavery S, Slade R (1995) Using allele frequency and phylogeny to define units for conservation and management. American Fisheries Society Symposium, 17, 249-262.

O'Corry-Crowe GM, Suydam RS, Rosenberg A, Frost KJ, Dizon AE (1997)

Phylogeography, population structure and dispersal patterns of the beluga whale

Delphinapterus leucas in the western Nearctic revealed by mitochondrial DNA. Molecular

Ecology, 6, 955-970.

Pichler FB, Dawson SM, Slooten E, Baker CS (1998) Geographic isolation of Hector's dolphin populations described by mitochondrial DNA sequences. Conservation Biology, 12,

676-682.

Pope LC, Sharp A, Moritz C (1996) Population structure of the yellow-footed rock-wallaby

Petrogale xanthopus (Gray, 1854) inferred from mtDNA sequences and microsatellite loci.

Molecular Ecology, 5, 629-640.

Ramsay EP (1888) Notes on the fauna of the Bellenden-Ker Ranges. Proceedings of the

Linnean Society of New South Wales, 2, 1295-1299.

Raymond M, Rousset F (1995) GENEPOP: Population genetics software for exact tests and ecumenicism. The Journal of Heredity, 86, 248-249.

Rice WR (1989) Analyzing tables of statistical tests. Evolution, 43, 223-225.

182 Rojas M (1992) The species problem and conservation: what are we protecting?

Conservation Biology, 6, 170-178.

Ryder OA (1986) Species conservation and systematics: the dilemma of subspecies. Trends in Ecology and Evolution, 1, 9-10.

Sambrook E, Fritsch F, Maniatis T (1989) Molecular Cloning: a Laboratory Manual, 2nd ed. Cold Spring Harbour Laboratory Press, New York.

Slatkin M (1995) A measure of population subdivision based on microsatellite allele frequencies. Genetics, 139, 457-462.

Swofford DL (1993) PAUP: Phylogenetic analysis using parsimony version 3.1.

Computer program distributed by the Illinois Natural History Survey.

Swofford DL, Selander RB (1981) BIOSYS-1: a FORTRAN program for the comprehensive analysis of electrophoretic data in population genetics and systematics. The

Journal of Heredity, 72, 281-283.

Tamura K (1992) Estimation of the number of nucleotide substitutions when there are strong transition-transversion and G+C-content biases. Molecular Biology and Evolution, 9, 678-

687.

183 Tate GHH (1952) Results of the Archbold expeditions. No. 66. Mammals of Cape York

Peninsula, with notes on the occurrence of rain forest in Queensland. Bulletin of the

American Museum of Natural History, 98, 563-616.

Taylor AC, Sherwin WB, Wayne RK (1994) Genetic variation of microsatellite loci in a bottlenecked species: the northern hairy-nosed wombat Lasiorhinus krefftii. Molecular

Ecology, 3, 277-290.

Vane-Wright RI, Humphries CJ, Williams PH (1991) What to protect? Systematics and the agony of choice. Biological Conservation, 55, 235-254.

Vogler AP, DeSalle R (1994) Diagnosing units of conservation management. Conservation

Biology, 8, 354-363.

Walker D, Ortí G, Avise JC (1998) Phylogenetic distinctiveness of a threatened aquatic turtle

(Sternotherus depressus). Conservation Biology, 12, 639-645.

Weir BS, Cockerham C (1984) Estimating F-statistics for the analysis of population structure.

Evolution, 38, 1358-1370

Worthington Wilmer J, Hall L, Barratt E, Moritz C (in press) Genetic structure and male mediated gene flow in the ghost bat (Macroderma gigas). Evolution.

184 Wolstenholme DR (1992) Animal mitochondrial DNA: structure and evolution. International

Review of Cytology, 141, 173-216.

Wright S (1951) The genetical structure of populations. Annals of Eugenetics, 1, 323-334.

Zhu D, Degnan S, Moritz C (1998) Evolutionary distinctiveness and status of the endangered

Lake Eacham rainbowfish (Melanotaenia eachamensis). Conservation Biology, 12, 80-93.

Zink RM, Kale HW (1995) Conservation genetics of the extinct dusky seaside sparrow

Ammodramus maritimus nigrescens. Biological Conservation, 74, 69-71.

Acknowledgments

I wish to thank L. Fumagalli for providing mtDNA primers before they were published, M.

Elphinstone and D. Jerry for instructing me in the use of TGGEs, and M. Heslewood for statistical help. I would also like to thank C. Belcher, S. Burnett, M. Jones, veterinary staff from Featherdale Wildlife Park and Taronga Zoo, the owners, managers, and staff at

Barrington Guest House, and others who assisted with the collection of samples. W. Greville provided assistance throughout all aspects of this study. C. Dickman, J. Ginsberg, L.

Bernatchez and two anonymous referees provided insightful critiques of the manuscript. This project was funded by The Winifred Violet Scott Trust, the Zoological Parks Board of NSW,

Barrington Guest House, State Forests of NSW, and Cassegrain Wines.

185 Chapter 6. Summary of Results and Conservation Implications for

Quolls

Introduction

The aim of this thesis was to provide a molecular analysis to assist in the conservation management of the large carnivorous marsupials within the genus Dasyurus. This series of studies is the first to provide an in-depth examination of the molecular ecology of quolls, and indeed, the first on any of the dasyurids that I am aware of. The four preceding chapters have shown the wide utility of highly variable genetic markers in the examination of evolutionary histories, levels of population variability, and differentiation between MUs and ESUs within species. These previous studies also highlight the potential application of molecular analyses to improve conservation management in quolls.

Summary of Results

As has been shown in a wide variety of other species, allozymes and other genetic markers often have little variability and provide limited information in comparison to microsatellite markers (e.g. Paetkau and Strobeck 1994; Estoup et al. 1998; Goodman 1998). Thus the identification and development of highly variable markers such as the microsatellites used in these studies (Chapter 2; Firestone 1999) was a necessary first step in elucidating the levels of diversity and differentiation within and between different populations and species of quolls.

The development of these microsatellite loci formed a substantial part of this project and these primers were the first such markers developed for any of the carnivorous marsupials within the family Dasyuridae.

186 Primer pairs for six highly polymorphic and informative microsatellite loci were developed which have the potential for a wide range of applications in further studies in quolls. These markers may also be useful in answering similar questions in other closely related species, such as the Tasmanian Devil, Sarcophilus harrisii, and in some other more distant members of the family such as Antechinus.

The results of control region phylogenetic analyses (Chapter 3; Firestone in press) emphasise the relationships among all species of quolls and finally bring some consensus to the question of quoll relationships. The findings presented here concur with the work of Krajewski et al. (1997) in establishing the close relationship of the western quoll, D. geoffroii, to the New Guinean species, D. spartacus and D. albopunctatus. Although the concurrence between these two data sets may be because they are both based on mitochondrial loci, the agreement between them is notable in light of the confusing history of phylogenetic studies relating to quolls over the last two or three decades. These results also highlight the importance of using more than one exemplar of each species when available, and confirms that species haplotypes form monotypic clusters, unlike in other closely related species such as the Sminthopsis.

However, a true comparison of mtDNA (Figures 11 and 12) and microsatellite (Figure 19) phylogenies is not possible. The differences in taxon sampling (individuals vs. populations), the use or not of outgroups (use of outgroups is possible in mtDNA phylogenies due to the conserved nature of the mtDNA locus; use of outgroups is difficult to do for phylogenies based on microsatellites due to the unconserved nature of microsatellite PCR primers), and the differing characterisitics of the loci (mtDNA phylogenies are based upon sequence data; microsatellite phylogenies are based upon frequency data) make direct comparisons of these phylogenies difficult.

The geographical structuring of mtDNA haplotypes within the two species where different localities were sampled (D. maculatus and D. hallucatus) show distinctly divergent lineages. This finding highlights the long-term evolutionary divisions within these species and is an important part of the process of defining conservation units. These results also show that the genetic distinctions between the bronze quoll, D. spartacus, and the western quoll, D. geoffroii, are low in comparison to some intraspecific comparisons (e.g. differences between Tasmanian and mainland tiger

187 quolls or between Queensland and Northern Territory northern quolls), leading to the question of where do we draw the line between specific and subspecific distinctions among quolls.

Examination of further loci and additional individuals may help elucidate this point.

The results presented on the variability and differentiation within and between populations and species of the four Australian quolls (Chapter 4; Firestone et al. submitted), highlights the usefulness of the microsatellite markers developed in Chapter 2. These markers were useful in defining differences at both the species and population levels and also provide the genetic basis for further population monitoring and other conservation applications. The major findings of this study show that 1) the microsatellites used here may prove to be a valuable tool in forensic analyses since they were useful for defining differences between species and populations and were highly diagnostic in assignment tests, 2) contrary to expectations based on the history of widespread population decline, western quolls have higher levels of diversity than any other species examined, and 3) most populations within each species should be treated as separate management units for conservation purposes based on differences in allele frequencies at microsatellite loci.

Analysis of population structure among tiger quolls (Chapter 5; Firestone et al. 1999) also highlights the utility of using independent genetic markers in determining conservation units.

Phylogenetic analyses indicate that there are reciprocally monophyletic groups within D. maculatus corresponding to a split between mainland and Tasmanian lineages. Similarly there are significant differences in allele frequencies

187 between the Tasmanian population and all other mainland populations indicating two distinct

ESUs. However, these ESUs do not correspond to the currently accepted subspecific categories based on morphological characters and geographical locations within D. maculatus, and indicate that the major split is between the Tasmanian populations of tiger quolls and those on the mainland. The Tasmanian populations are quite distinct, genetically but not morphologically, from those on the mainland. However, the subspecies D. m. gracilis, is not distinct genetically, but is distinct morphologically from the southern subspecies D. m. maculatus. Thus it is the combination of genetic and morphological data that define subspecies within this species, and the Tasmanian populations will need to be elevated to the subspecies status.

Furthermore, there were significant differences in allele frequencies amongst most mainland populations of tiger quolls examined. These results indicate that there are many functionally isolated units within this species and that these units should be treated as distinct MUs according to the criteria of Moritz (1994). Analyses of two different genetic systems (mtDNA control region and nuclear microsatellites) provide different results, however, when highly localized populations are examined, providing a clue as to the population structure and social biology of tiger quolls in the Barrington Tops region. Analysis of microsatellites indicate that the Barrington populations were not subdivided and should be treated as one MU. However, analysis of mtDNA indicates that there are significantly subdivided populations within this area corresponding to matrilineal lineages within valley systems. Since microsatellites are biparentally inherited but mtDNA is only maternally inherited, the differences observed using the two markers may be attributed to sex-biased dispersal patterns. Interestingly,

188 this finding concurs with preliminary radiotracking data (not presented here) which shows that females have small localized home ranges within valleys, while males have much wider ranges, traversing valley systems.

Conservation Implications of Relevance to the Management of Quolls

There are a number of possible actions pertinent to quolls that agencies or managers might attempt as a means of conserving each of these species. For instance, there have been documented attempts to breed Tasmanian and mainland tiger quolls and release the progeny onto the mainland (Fleay 1940; Fleay 1948). If the north Queensland tiger quoll populations continue to decline, there may be proposals to supplement these populations with captive bred or wild caught individuals from other subpopulations within this region. Likewise, in future, managers might wish to translocate northern quolls between different regions as a means of supplementing declining populations. Currently, there are also proposals to reintroduce eastern quolls from Tasmania onto the mainland. The application of genetic data to the conservation of quolls can provide much needed information to help guide future management plans. For instance, a number of recommendations for the conservation of quolls emerge from the genetic analyses presented here.

· Distinct phylogenetic lineages within species indicate long term evolutionary splits that

potentially may lead to speciation events. Tasmanian and mainland tiger quolls form their

own separate ESUs and should not be bred together in any captive breeding programs.

Similarly, the Tasmanian tiger quolls should not be translocated for release onto the

mainland and mainland tiger quolls should not be translocated to Tasmania.

189 · Although the mainland populations of tiger quolls lack phylogenetic distinctions, most of these populations are demographically isolated and currently operate as functionally separate units (with the possible exception of the Barrington Tops populations). Analysis of population subdivision and geographic distance supports this finding and indicates that dispersal distances are quite short within this species. Thus animals should not routinely be transported between MUs unless population numbers drop to such levels that supplemental translocations are required. However, it must also be noted that in some instances the genetic differences between these populations are relatively minor and could readily be outweighed by the cost of separate conservation and the threat of inbreeding. Therefore a balanced judgement must prevail.

· Most populations within the other quoll species examined also are demographically isolated

and should be considered as separate MUs. Thus the three Tasmanian populations of

eastern quolls that were examined are all currently differentiated based on allele frequencies

and, again, supplementation of individuals from one population to another should only be

done if population numbers drop to low numbers. Similarly, although only a few

populations of northern quolls were available for this study, the Northern Territory and

Queensland populations of northern quolls are separate MUs and possibly separate ESUs,

thus these two populations should not be admixed.

· Continued genetic monitoring of the captive colony at Perth Zoo and translocated

populations of western quolls will be necessary as a means of determining inbreeding or

founder effects in these populations.

190 Areas for Further Study

The results presented here provide a blueprint for the conservation genetics of quolls. These studies are only preliminary, however, and open up a host of other questions. Further room exists for expanded genetic studies into: 1) a broad scale examination of the phylogeographic population structure of D. hallucatus as a means of elucidating conservation units/subspecific status of this species, 2) an examination of population variability and differentiation within and among populations of both of the New Guinean species, D. spartacus and D. albopunctatus,

3) an examination of parentage or paternity exclusion in populations of quolls where extensive field data exists, 4) a metapopulation study of gene flow and genetic differentiation between the remaining subpopulations of D. m. gracilis, 5) follow up studies on the level of genetic variability remaining in translocated populations of western quolls, D. geoffroii, to assess founder effects or levels of inbreeding, and 6) further studies using both additional loci and individuals to elucidate the degree of differentiation between D. spartacus and D. geoffroii.

Furthermore, despite the utility of both nuclear microsatellites and the mtDNA control region in defining conservation units within quolls, there are still operational difficulties in defining both

ESUs and MUs. For example, if two populations were significantly different at one locus, but

the overall subdivision value (either RST, FST, or FST) measured over all loci was not significantly different, would these two populations be considered as separate MUs? Similarly, if two sequences are reciprocally monophyletic due to one base pair difference out of 1000 bases, would these be diagnostic of separate

191 ESUs? Clearly there is a need for more theoretical work and better practical guidelines in defining conservation units.

References

Estoup A, Rousset F, Michalakis Y, Cornuet J-M, Adriamanga M, Guyomard R (1998)

Comparative analysis of microsatellite and allozyme markers: a case study investigating microgeographic differentiation in brown trout (Salmo trutta). Molecular Ecology, 7, 339-

353.

Firestone K (1999) Isolation and characterization of microsatellites from carnivorous marsupials (Dasyuridae:Marsupialia). Molecular Ecology, 8, 1084-1086.

Firestone K, Elphinstone M, Sherwin B, Houlden B (1999) Phylogeographical population structure of tiger quolls Dasyurus maculatus (Dasyuridae:Marsupialia), an endangered carnivorous marsupial. Molecular Ecology, 8, 1613-1626.

Firestone KB (in press) Phylogenetic relationships among quolls revisited: the mtDNA control region as a useful tool. Journal of Mammalian Evolution.

Firestone KB, Sherwin WB, Houlden BH, Geffen E (submitted) Variability and differentiation: microsatellites in the genus Dasyurus and conservation implications for the large Australian carnivorous marsupials. Conservation Genetics.

192 Fleay D (1940) Breeding of the tiger-cat. Victorian Naturalist, 56, 159-163.

Fleay D (1948) Australia's marsupial tiger cat: a creature of increasing rarity and extreme pugnacity. Wildlife Australian Nature Magazine, 10, 459-464.

Goodman SJ (1998) Patterns of extensive genetic differentiation and variation among

European harbor seals (Phoca vitulina vitulina) revealed using microsatellite DNA polymorphisms. Molecular Biology and Evolution, 15, 104-118.

Krajewski C, Young J, Buckley L, Woolley PA, Westerman M (1997) Reconstructing the evolutionary radiation of Dasyurine marsupials with cytochrome b, 12S rRNA, and protamine

P1 gene trees. Journal of Mammalian Evolution, 4, 217-236.

Moritz C (1994) Defining 'Evolutionarily Significant Units' for conservation. Trends in

Ecology and Evolution, 9, 373-375.

Paetkau D, Strobeck C (1994) Microsatellite analysis of genetic variation in black bear populations. Molecular Ecology, 3, 489-495.

193 Appendix 1. Cloned microsatellite sequences. Areas underlined are the forward primer sequence, areas in bold are the reverse primer sequence. Regions where the sequence could not be read are indicated by (n)n .

Locus 1.3 caattgggga attgatgaac aagacatagc gtttgattgt gatatagtat tattttacta taagaaataa tgaaaggttg gtttcagaaa cacacacaca cacacacaca cacacacaca tcacatgcac agtaagactt atatgaaatg atgctaagtg acctaagcat aaccaggaaa tcattgtaca tggtgacatc tatatttcaa taagaatcaa ctgtgaaaga ctta

Locus 3.1.2 cttggtgttg tcccattcta tatccagaat ggaaggaaac ttcacaagtg tcgaaacaat aaattaattt acctatgaat aaaatagatg tagtcattct aaacaaaagg taaatctaga gggaggtggt agccagggag aacatacaca cacacacaca cacacacaca cacacacaca ctgcaagtta gttcactttc caacaacaga tgagtcatta atcag

Locus 3.3.1 ccagcccttg agtcttgaga tttgactcca gatgtgtgtt gtgtggtgtg gtgtgtctat gtgtgtgtgt gtgtgtgtgt gtgtgtgtgt gtgtgtgtga gacctttatg tccttttgga gggaaactcc tggggtggta tgtgggcaga tagagggagt agtgtgtgtg tagtaatgat gaaaaggaaa gccaccttga gcacgattcc ctctccagat cctccagaag cctaccctga tcag

Locus 3.3.2 tttattaaaa gcccagttgn gtggatgact tgctcattcc taaaggtact aaatagcaga gactcgatcc acaccaagat cttggatacc caactagg(n)nt aacacacaca cacacacaca cacacacaca cacacacaca cacatatctc atagaatctt cccaggtaat aaaggctact cagaggaatg tgatcctgaa ttgttagagg tanaggttac tatctggg(n)na gttgggagga cgaccttggt ctctgcttaa aatgggatcc ttct

Locus 4.4.2 ccatcagctg gggcatgctc tgggaaatcc aagctcattt tagtattaag attaataaag agaataatga atctgcaaac acacacacac acacacacac acacacacac acacacagca agtcacgaga tgcattccag agttgattac cgtacaagca gcttgggagt catggcttct gttttggatt attcatgctg gaattcccgg

Locus 4.4.10 ccagagctat cttaataagt gagaatgcta gatttcactc ccatttgttt cctcattttc cttctgattt tgacttagtg cttttctttt tagttttaaa tttcatgatc aaagtatata agatcagcag tggggcagtg tgtcagaaaa ggagtgtgtg tgtgtgtgtg tgtgtgtgtg tgtgtgtgtg tgtgtgtgtg tctgtgtgtg tgaagggaga agtatgcaag gggagtgaga tcatagaatc agaaaacagt tccagaaatg tgaggtgagg tgagctggct gccccagatc acacaaagtt acccatatcc ccgactctta gttctgtgct ccttgcactc caccatgatc tctacaatc

195 Appendix 2. Voucher data for all samples used in this study. All samples were collected from live animals, road killed animals, or museum specimens. All procedures were non-lethal and collected under appropriate permits from the various State wildlife agencies and under ACEC permit number 3D798 issued b~ the ZPB ofNSW. 2 Species/ ID# Age/ Date Location collected Collector/ 3 Pop1 Sex collected Institution4 Dasr_urus maculatus co NM.E1 M Jan.21,93 Copeland, NSW J. Giles co NM.E2 NM Apr. 22,93 Copeland, NSW K. Firestone (KBF) co NM.E3 J/F May 5, 93 Scone Road, Copeland, NSW KBF co NM.E4 J/F Jun.28,93 Copeland, NSW KBF co NM.E5 NF Jun.28, 93 Copeland Tops State Forest, KBF NSW co NM.E6 NM Jun. 30,93 Scone Road, Copeland, NSW KBF co NM.E7 J/M Jul. 2, 93 Copeland Tops SF, NSW KBF co NM.L1 NM May 5, 93 Bowman Map: 9234-2-S J. Griffith S: 6460925 E: 385575 co NM.L 11 NM Jun.26,95 Bowman Map: 9234-2-S J. Griffith S:6461800 E: 382050 BG NM.E9 NF Nov. 18, 93 Barrington Guest House, KBF Salisbury, NSW (BGH) BG NM.E10 NM Nov. 20,93 BGH KBF BG NM.E11 J/M Nov. 20,93 BGH KBF BG NM.E12 J/M Dec.20,93 BGH KBF BG NM.E13 J/F Dec. 21, 93 BGH KBF BG NM.E14 NM May 15,94 BGH KBF BG NM.E15 NM May 16,94 BGH KBF BG NM.E17 NM Ma~ 19,94 BGH KBF BG NM.E18 NM May 21,94 BGH KBF BG NM.E19 NM Aug 12,94 BGH KBF BG NM.E20 J/M Dec. 10, 94 BGH KBF BG NM.E24 J/M Jan 15, 95 BGH KBF BG NM.E33 NM Oct. 12,95 BGH KBF BG NM.E36 NM Jul~ 27, 96 BGH KBF BG NM.E37 NF July 28, 96 BGH KBF BG NM.E38 NM Jul~ 31, 96 BGH KBF BG NM.L3 NM Jul. 18, 94 BGH B. Lewis BG NM.T6 py Jul~ 28, 96 BGH KBF BG NM.T7 py Jul~ 28, 96 BGH KBF BG NM.T8 py Jul~ 28,96 BGH KBF BG NM.T9 py July 28,96 BGH KBF BG NM.T10 PY Jul~ 28,96 BGH KBF BG NM.T11 py Jul~ 28,96 BGH KBF CH NM.E16 J/F May 17,94 Chichester State Forest, NSW KBF {CSF) CH NM.E21 J/M Jan. 7,95 CSF KBF CH NM.E22 NF Jan.8,95 CSF KBF CH NM.E23 J/M Jan.8,95 CSF KBF CH NM.E25 NF CSF KBF CH NM.E26 NM Apr. 4, 95 CSF KBF CH NM.E28 NF Sept. 5, 95 CSF KBF CH NM.E29 NF SeEt. 7, 95 CSF KBF CH NM.E30 NM Sept. 7, 95 CSF KBF CH NM.E32 NM Oct. 9, 95 CSF KBF CH NM.E40 NM Aug. 10,96 CSF KBF CH NM.E41 NF Oct. 6, 96 CSF KBF CH NM.L4 NM Jun. 7,94 Chichester Map: 9233-IV-S D. Longbottom 8:6431600 E:362800. CH NM.T1 PY/M Sept. 7, 95 CSF KBF

196 2 Species/ ID# Age/ Date Location collected Collector/ 4 Pop1 Sex3 collected lnstitution CH NM.T2 PY/F Sept. 7, 95 CSF KBF CH NM.T3 PY/F Sept. 7, 95 CSF KBF CH NM.T4 PY/M Sept. 7, 95 CSF KBF CH NM.T5 PY/F Sept. 7, 95 CSF KBF BT NM.E8 AIM Oct. 20, 93 Nowendoc River, NSW KBF BT NM.E27 J/F Feb.4,95 Gresford Map: 9233-3-S KBF S:6414050 E:373750 BT NM.L9 J/F Feb.4,95 Gresford Map: 9233-3-S B. Read S:6414050 E:373750 BT NM.E31 AIM? Sept. 6, 96 Camberwell Map: 9133-111-S J. Burnet S:6409300 E:329650 BT NM.E35 JIM May6, 96 Putty Road, 5 km S of KBF Singleton, NSW. BT NM.E39 AIM May 26,96 Bandon Grove, NSW B. Dowling BT NM.L2 AIF 1990 (?) Chichester Map: 9233-IV-S R.Darr S: 6440650 E: 361175 BT NM.L7 AIM Oct. 15, 94 Allynbrook Map: 9233-3-S J. Byron S: 6418800 E: 381125 BT NM.L8 AIM Dec. 11,94 Moonan Brook Map: 9134-2-S F. Craven 8:63750 E:52400 BT NM.L 10 AIM Ma~ 1, 95 Uffington State Forest, NSW J. Titmarsh BT NM.L 13 AIM Mar. 18, 96 Bulahdelah Map: M. Murray S: 6450700 E: 410200 BS NM.E42 M Badga State Forest, NSW C. Belcher BS NM.E43 M Badga State Forest, NSW C. Belcher BS NM.E44 AIM Aug. 25,96 Badga State Forest, NSW C. Belcher BS NM.E45 AIM Aug.29,96 Badga State Forest, NSW C. Belcher BS NM.E46 AIM Sept. 5, 96 Badga State Forest, NSW C. Belcher BS NM.E47 AIM Sept. 5, 96 Badga State Forest, NSW C. Belcher Gl NM.B1 AIM Feb.24,93 zoo bred, origin Glenn Innes R. Dickens/FWP Gl NM.B2 AIM Feb.24,93 zoo bred, origin Glenn Innes R. Dickens/FWP Gl NM.B3 AIF Jul. 21,93 zoo bred, origin Glenn Innes L. Vogelnest!TZ/ FWP Gl NM.Mu 1 M zoo bred, origin Glenn Innes M. Westerman/LU/ Healesville/FWP Gl NM.E34 AIM Ma~7,96 zoo bred, origin Glenn Innes KBF TT TM.E1 Tasmania M. Jones/UTAS TT TM.E2 Tasmania M. Jones/UTAS TT TM.E3 Tasmania M. Jones/UTAS TI TM.E4 Tasmania M. Jones/UTAS TT TM.E5 Tasmania M. Jones/UTAS TT TM.E6 Tasmania M. Jones/UTAS TT TM.E7 Tasmania M. Jones/UTAS TT TM.E8 Tasmania M. Jones/UTAS TT TM.E9 AIM Nov. 25,92 northwest Tasmania A. Kelly/TWP TT TM.E10 AIM Nov. 25,92 west central Tasmania A. KeUy/TWP TT TM.E11 AIF Nov. 25,92 north central Tasmania A. Kelly/TWP WY TM.M1 M Nov. 12, 11 Wynyard, Tasmania MoVC6099 Lat: 41°00' S Long: 145°44' E WY TM.M2 M Oct. 19, 11 W~n~ard, Tasmania MoVC6100 WY TM.M3 M Feb.22, 12 W~n~ard, Tasmania MoVC6101 WY TM.M4 M June 27, 12 W~n~ard, Tasmania MoVC6102 WY TM.M5 M Jul~ 11, 12 W~nyard, Tasmania MoVC6103 WY TM.M6 M Jul~ 11, 12 W~ny_ard, Tasmania MoVC6104 WY TM.M7 W~ny_ard, Tasmania MoVC6105

197 2 Species/ ID # Age/ Date Location collected Collector/ 3 4 Pop1 Sex collected lnstitution WY TM.M8 M July 26, 12 Wy_ny_ard, Tasmania MoV C6106 WY TM.M9 F Jan 8, 14 Wyny_ard, Tasmania MoV C6107 WY TM.M10 M Feb.6, 14 Wynyard, Tasmania MoVC6108 WY TM.M11 F June 4, 14 Wynyard, Tasmania MoVC6109 WY TM.M12 M June 19, 14 Wynyard, Tasmania MoV C6110 WY TM.M13 M July 2, 14 Wy_nyard, Tasmania MoV C6111 WY TM.M14 M July_ 9, 14 Wy_nyard, Tasmania MoV C6112 WY TM.M15 M July_ 23, 14 Wynyard, Tasmania MoV C6113 WY TM.M16 F July 23, 14 Wynyard, Tasmania MoVC6114 WY TM.M17 M July 23, 14 Wynyard, Tasmania MoVC6115 WY TM.M18 M July 23, 14 Wyny_ard, Tasmania MoVC6116 WY TM.M19 M July 22, 14 Wy_nyard, Tasmania MoV C6117 WY TM.M20 M July 22, 14 Wynyard, Tasmania MoV C6118 WY TM.M21 F Nov. 14, 13 Wynyard, Tasmania MoV C6119 WY TM.M22 M Aug. 7, 14 Wy_ny_ard, Tasmania MoVC6120 WY TM.M23 F Nov. 14, 13 Wy_nyard, Tasmania MoV C6121 WY TM.M24 F May_ 7, 14 Wy_ny_ard, Tasmania MoVC6122 WY TM.M25 M July_ 2, 14 Wy_nyard, Tasmania MoVC6123 WY TM.M26 M June 4, 14 Wy_ny_ard, Tasmania MoV C6124 WY TM.M27 M June 4, 14 Wy_ny_ard, Tasmania MoVC6125 WY TM.M28 M July_ 2, 14 Wyny_ard, Tasmania MoVC6126 WY TM.M29 M June 29,22 Wy_ny_ard, Tasmania MoVC6127 WY TM.M30 M July 2, 14 Wy_nyard, Tasmania MoVC6128 WY TM.M31 M May 28, 14 Wyny_ard, Tasmania MoVC6129 WY TM.M32 Wynyard, Tasmania MoVC6130 WY TM.M33 Wy_ny_ard, Tasmania MoVC6131 WY TM.M34 Wynyard, Tasmania MoVC6132 WY TM.M35 Wy_ny_ard, Tasmania MoVC6133 MW QM.E1 F Aug. 11, 93 Mt. WindsorTableland, Qld KBFI S. Burnet MW QM.E2 F Au9. 15,93 Mt. Windsor Tableland, Qld KBFI S. Burnet MW QM.E3 M Aug. 13,93 Mt. Windsor Tableland, Qld KBFI S. Burnet MW QM.E4 F Aug. 15,93 Mt. Windsor Tableland, Qld KBFI S. Burnet MW QM.ES M Aug. 14,93 Mt. Windsor Tableland, Qld KBFI S. Burnet MW QM.E6 M Au9. 12,93 Mt. Windsor Tableland, Qld KBF/ S. Burnet MW QM.E7 M Aug. 12,93 Mt. Windsor Tableland, Qld KBFI S. Burnet MW QM.E8 M Aug.13, 93 Mt. Windsor Tableland, Qld KBFI S. Burnet MW QM.E9 M Au9. 12,93 Mt. WindsorTableland, Qld KBF/ S. Burnet MW QM.E10 M Aug. 14,93 Mt. Windsor Tableland, Qld KBF/ S. Burnet MW QM.E11 F Au9.11,93 Mt. Windsor Tableland, Qld KBF/ S. Burnet MW QM.L1 F Aug. 13,93 Mt. Windsor Tableland, Qld KBF/ S. Burnet SB VM.L2 AIM Jun.22,86 Suggan Buggan. Vic R. Morgan/ DCNR, S: 36° 57' E: 148° 22' Vic. SB VM.E3 M Suggan Buggan, Vic C. Belcher SB VM.E4 F Suggan Buggan, Vic C. Belcher Dasr_urus viverrinus TE TV.E1 Tasmania M. Jones/UTAS TE TV.E2 Tasmania M. Jones/UTAS TE TV.E3 Tasmania M. Jones/UTAS TE TV.E4 Tasmania M. Jones/UTAS TE TV.ES Tasmania M. Jones/UTAS TE TV.E6 Tasmania M. Jones/UTAS TE TV.E7 Tasmania M. Jones/UTAS TE TV.E8 Tasmania M. Jones/UTAS TE TV.E9 Tasmania M. Jones/UTAS TE TV.E10 Tasmania M. Jones/UTAS

198 2 Species/ 10 # Age/ Date Location collected Collector/ 3 4 Pop1 Sex collected lnstitution TE TV.E11 AJF Nov. 25,92 south Tasmania A. Kell~nwP TE TV.E12 AiM Nov. 25, 92 east central Tasmania A. KellynwP TE TV.E13 AiM Nov. 25, 92 southeast Tasmania A. KellynwP TE TV.E14 AiM Jut~ 8, 94 Lake Leake, Tasmania KBF VB TV.E15 Apr. 15, 95 Vale of Belvoir, Central Tas. D. Moyle/UlAS VB TV.E16 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E17 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Moyle/UTAS VB TV.E18 Apr. 17, 95 Vale of Belvoir, Central las. D. Moyle/UTAS VB TV.E19 Apr. 16, 95 Vale of Belvoir, Central las. D. Mo~le/UTAS VB TV.E20 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Moyle/UTAS VB TV.E21 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Moyle/UTAS VB TV.E22 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E23 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E24 Apr. 15, 95 Vale of Belvoir, Central Tas. D. Moyle/UTAS VB TV.E25 Apr. 13, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E26 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E27 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E28 Apr. 13, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E29 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E30 Apr. 14, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E31 Apr. 15,95 Vale of Belvoir, Central Tas. D. Moyle/UTAS VB TV.E32 Apr. 14,95 Vale of Belvoir, Central Tas. D. Moyle/UTAS VB TV.E33 Apr. 15, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E34 Apr. 15, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS VB TV.E35 Apr. 16, 95 Vale of Belvoir, Central Tas. D. Mo~le/UTAS GL TV.M1 M May 13,61 Gladstone, N. E. Tasmania QVM 1963/1/283 40° 58'S 148° 01' E GL TV.M2 M Ma~ 13,61 Gladstone, Tasmania QVM 1963/1/284 GL TV.M3 F Ma~ 13,61 Gladstone, Tasmania QVM 1963/1/286 GL TV.M4 M Oct. 7, 63 Gladstone, Tasmania QVM 1963/1/288 GL TV.M5 F Jul~ 5, 64 Gladstone, Tasmania QVM 1964/1/1 07 GL TV.M6 M Jul~ 5, 64 Gladstone, Tasmania QVM 1964/1/1 09 GL TV.M7 F June 6, 64 Gladstone, Tasmania QVM 1964/1/111 GL TV.M8 M July 9, 64 Gladstone, Tasmania QVM 1964/1/116 GL TV.M9 M July 20,64 Gladstone, Tasmania QVM 1964/1/125 GL TV.M10 M July20, 64 Gladstone, Tasmania QVM 1964/1/126 GL TV.M11 M July20, 64 Gladstone, Tasmania QVM 1964/1/127 GL TV.M12 M Jul~20, 64 Gladstone, Tasmania QVM 1964/1/128 GL TV.M13 F Jul~ 21,64 Gladstone, Tasmania QVM 1964/1/129 GL TV.M14 M July 22,64 Gladstone, Tasmania QVM 1964/1/136 GL TV.M15 M July23, 64 Gladstone, Tasmania QVM 1964/1/141 GL TV.M16 M July 23,64 Gladstone, Tasmania QVM 1964/1/142 GL TV.M17 M July 23,64 Gladstone, Tasmania QVM 1964/1/143 GL TV.M18 M July23, 64 Gladstone, Tasmania QVM 1964/1/144 GL TV.M19 M July 23,64 Gladstone, Tasmania QVM 1964/1/145 GL TV.M20 M July23, 64 Gladstone, Tasmania QVM 1964/1/147 GL TV.M21 M July 23,64 Gladstone, Tasmania QVM 1964/1/148 GL TV.M22 F July 23,64 Gladstone, Tasmania QVM 1964/1/149 GL TV.M23 M July 24,64 Gladstone, Tasmania QVM 1964/1/152 GL TV.M24 M July 24,64 Gladstone, Tasmania QVM 1964/1/153 GL TV.M25 M July24, 64 Gladstone, Tasmania QVM 1964/1/154 GL TV.M26 F July 24,64 Gladstone, Tasmania QVM 1964/1/155 GL TV.M27 M July 24, 64 Gladstone, Tasmania QVM 1964/1/159 GL TV.M28 M July 24, 64 Gladstone, Tasmania QVM 1964/1/162 GL TV.M29 M July 25,64 Gladstone, Tasmania QVM 1964/1/163

199 2 Species/ 10# Age/ Date Location collected Collector/ Pop1 Sex3 collected lnstitution4 GL TV.M30 F July 25, 64 Gladstone, Tasmania QVM 1964/1/164 GL TV.M31 F July 25, 64 Gladstone, Tasmania QVM 1964/1/165 GL TV.M32 M July 25,64 Gladstone, Tasmania QVM 1964/1/168 GL TV.M33 F July 25,64 Gladstone, Tasmania QVM 1964/1/169 GL TV.M34 M July 25,64 Gladstone, Tasmania QVM 1964/1/171 GL TV.M35 M July 25,64 Gladstone, Tasmania QVM 1964/1/172 GL TV.M36 F July 25,64 Gladstone, Tasmania QVM 1964/1/173 GL TV.M37 M July 26,64 Gladstone, Tasmania QVM 1964/1/174 GL TV.M38 M July 26,64 Gladstone, Tasmania QVM 1964/1/175 GL TV.M39 M July 26,64 Gladstone, Tasmania QVM 1964/1/176 GL TV.M40 M July 26, 64 Gladstone, Tasmania QVM 1964/1/177 GL TV.M41 M July 26,64 Gladstone, Tasmania QVM 1964/1/181 GL TV.M42 M July 26,64 Gladstone, Tasmania QVM 1964/1/182 GL TV.M43 F July 26,64 Gladstone, Tasmania QVM 1964/1/183 GL TV.M44 M July 26,64 Gladstone, Tasmania QVM 1964/1/185 GL TV.M45 M July 27,64 Gladstone, Tasmania QVM 1964/1/189 GL TV.M46 F July 27,64 Gladstone, Tasmania QVM 1964/1/190 GL TV.M47 F July 28,64 Gladstone, Tasmania QVM 1964/1/193 GL TV.M48 F July 28,64 Gladstone, Tasmania QVM 1964/1/194 GL TV.M49 F July 28,64 Gladstone, Tasmania QVM 1964/1/195 GL TV.M50 M July 29,64 Gladstone, Tasmania QVM 1964/1/204 GL TV.M51 F July 29,64 Gladstone, Tasmania QVM 1964/1/205 GL TV.M52 M July 30,64 Gladstone, Tasmania QVM 1964/1/209 GL TV.M53 M July 30,64 Gladstone, Tasmania QVM 1964/1/210 GL TV.M54 M July 30,64 Gladstone, Tasmania QVM 1964/1/211 GL TV.M55 M July 24,64 Gladstone, Tasmania QVM 1964/1/242 GL TV.M56 M June 27,65 Gladstone, Tasmania QVM 1965/1/143 GL TV.M57 M Gladstone, Tasmania QVM 1965/1/237 GL TV.M58 M May 13,61 Gladstone, Tasmania QVM 1982/1/305 ST W.M1 M May4, 11 Studley Park, Victoria MoVC6058 Lat: 37°48' S Long: 145° 01' E ST W.M2 F May 11, 11 Studley Park, Victoria MoVC6061 ST W.M3 M May 18, 11 Studley Park, Victoria MoVC6062 ST W.M4 F May 18, 11 Studley Park, Victoria MoVC6063 ST W.M5 M May 19, 11 Studley Park, Victoria MoVC6064 ST W.M6 M June 2, 11 Studley Park, Victoria MoVC6065 ST W.M7 F June 8, 11 Studley Park, Victoria MoVC6066 ST W.M8 F June 8, 11 Studley Park, Victoria MoVC6067 ST W.M9 M May 19, 11 Studley Park, Victoria MoVC6068 ST W.M10 F May 19, 11 Studley Park, Victoria MoVC6069 ST W.M11 M May23, 11 Studley Park, Victoria MoVC6070 ST W.M12 M May 23, 11 Studley Park, Victoria MoVC6071 ST W.M13 M June 2, 11 Studley Park, Victoria MoVC6072 ST W.M14 M Feb.11, 14 Studley Park, Victoria MoVC6074 ST W.M15 F Feb. 11, 14 Studley Park, Victoria MoVC6075 ST W.M16 M Feb. 11, 14 Studley Park, Victoria MoVC6076 ST W.M17 F Feb. 17, 14 Studley Park, Victoria MoVC6077 ST W.M18 M Feb. 15, 17 Yarra Bend, Victoria MoVC6078 Lat: 37° 48' S Long: 145° 01' E ST W.M19 F Jan 31, 17 Kew, Victoria MoVC6079 Lat: 37o 48'S Long: 145° 01' E ST W.M20 M Mar. 29, 17 Yarra Bend, Victoria MoVC6080 Lat: 37o 48'S long: 145° 01' E ST W.M21 M June 2, 11 Studley Park, Victoria MoVC6089 NE NV.M1 M Oct. 1887 NSW AM105

200 2 Species/ 10# Age/ Date Location collected Collector/ 4 Pop1 Sex3 collected lnstitution NE NV.M2 F NSW AM642 NE NV.M3 M Jul. 1, 1895 Colo Vale, NSW AM970 Lat: 34°24'8 Long: 150°29'E NE NV.M4 M Jul.1, 1895 Colo Vale, N8W AM971 Lat: 34°24'8 Long: 150°29'E NE NV.M5 M Jul10, 1895 Colo Vale, N8W AM973 Lat: 34°24'8 Long: 150°29'E NE NV.M6 F Sept 5, 1895 Goulburn, NSW AM 1002 Lat: 34°45'8 Long:149°43'E NE NV.M7 M Jan. 18, 1898 Lawson, N8W AM 1264 Lat: 33°43'8 Long: 150°26'E NE NV.M8 M Jun.9, 1899 Mosman, NSW AM 1433 Lat: 33°50'8 Long: 151 °15'E NE NV.M9 F Nov. 1, 1899 Kempsey, N8W AM 1477 Lat: 31°05'8 Long:152°50'E NE NV.M10 M Feb. 13, 1900 Mosman, N8W AM 1491 Lat: 33°50'8 Long:151°15'E NE NV.M11 F Jul. 30, 1900 Mosman, N8W AM 1517 Lat: 33°50'8 Long:151°15'E NE NV.M12 M Nov. 18,05 Waverley, N8W AM 1817 Lat: 40°15'8 Long: 151 °24'E NE NV.M13 M NSW AM 1902 NE NV.M14 M Mar. 31, 08 Lindfield, NSW AM 1938 Lat: 33°47'8 Long:151°10'E NE NV.M15 F Nov. 24, 15 Cremorne Point, NSW AM 2600 Lat: 23°06'8 Long:47° 02'E NE NV.M16 Jun. 16,21 Caught in Taronga Zoo AM 2972 Lat: 33°42' S Long:146°20'E NE NV.M17 M Aug. 9, 27 Vaucluse, NSW AM4106 Lat: 33°52'8 Long:151°17'E NE NV.M18 M Jan. 11,29 Vaucluse, NSW AM4484 Lat: 33°52'5 Long:151°17'E NE NV.M19 M NSW AM 5093 NE NV.M20 M Apr. 15, 32 Vaucluse, NSW AM 5269 Lat: 33°52'8 Long:151°17'E NE NV.M21 F Aug. 18, 38 Pennant Hills, NSW AM6524 Lat: 33°44'8 Long:151°04'E NE NV.M22 M Ma~8,39 Rose Ba~. N8W AM6604 NE NV.M23 F NSW AM 6625 NE NV.M24 F May27, 48 Vaucluse, NSW AM 7389 Lat:33°52'S Long:151°17'E NE NV.M25 F Aug. 20,63 Vaucluse, NSW AM 8382 Lat: 33°52'8 Long:151 °17'E Dasr_urus hallucatus KA NTH.L1 AIM MayS, 94 S: 12° 43' E: 132° 25' M. Oak.wood/ANU KA NTH.L2 AIM Jun.2,94 S: 12° 43' E: 132° 25' M. Oakwood/ANU KA NTH.L3 AIM Jun.2,94 S: 12° 42' E: 132° 25' M. Oakwood/ANU KA NTH.L4 AIM Jun.2,94 S: 12° 38' E: 132° 35• M. Oakwood/ANU KA NTH.L5 AIM Jun.2,94 S: 12° 44' E: 132° 23' M. Oakwood/ANU KA NTH.L6 AIM Jun.4,94 S: 12° 43' E: 132° 24' M. Oakwood/ANU KA NTH.L7 AIM Jun. 7,94 S: 12° 44' E: 132° 20' M. Oakwood/ANU KA NTH.L8 AIM Jun. 10,94 S: 12° 42' E: 132° 27' M. Oakwood/ANU KA NTH.L9 AIM Jun. 14, 94 S: 12° 39' E: 132° 43' M. Oakwood/ANU KA NTH.L 10 AIM Jun. 15, 94 S: 12° 38' E: 132° 38' M. Oakwood/ANU KA NTH.L 11 AIM Jun. 15,94 S: 12° 38' E: 132° 43' M. Oakwood/ANU

201 2 Species/ 10# Age/ Date Location collected Collector/ 3 4 Pop1 Sex collected Institution KA NTH.L 12 AIM Jun.29, 94 S: 12° 38' E: 132° 35' M. Oakwood/ANU KA NTHL.13 AIM Jul. 25, 94 S: 12° 38' E: 132° 38' M. Oakwood/ANU KA NTH.L 14 AIF Sept. 18, 94 S: 12° 38' E: 132° 34' M. Oakwood/ANU KA NTH.L 15 AIF Nov. 15,94 S: 12° 35' E: 132° 19' M. Oakwood/ANU KA NTH.E1 M Apr. 27,95 S: 12° 38• E: 132° 39' M. Oakwood/ANU KA NTH.E2 M Apr. 4, 95 S: 12° 38' E: 132° 39' M. Oakwood/ANU KA NTH.E3 M Apr. 2, 95 S: 12° 38' E: 132° 35' M. Oakwood/ANU KA NTH.E4 M Jun. 16, 95 S: 12° 38' E: 132° 35' M. Oakwood/ANU KA NTH.ES Jun. 94 Near Jabiru. M. Oakwood/ANU KA NTH.E6 M May 31, 94 S: 12° 42' E: 132° 26' M. Oakwood/ANU KA NTH.E7 May 27,94 S: 12° 43' E: 132° 24' M. Oakwood/ANU KA NTH.E8 M May 3, 94 S: 12° 38' E: 132° 38' M. Oakwood/ANU KA NTH.E9 F May 2, 94 S: 12° 44' E: 132° 22' M. Oakwood/ANU KA NTH.E10 F Sept. 13, 93 S: 12° 42' E: 132° 25' M. Oakwood/ANU KA NTH.E11 AIM May 10, 95 S: 12° 39' E: 132° 50' M. Oakwood/ANU AT QH.E1 J/M Feb. 12,94 Cape York Peninsula, Qld L. Leong/ AT QH.E2 A/F Dec. 11,95 Atherton Tablelands, Qld L. Pope/UQ AT QH.E3 A/F Dec. 12,95 Atherton Tablelands, Qld L. Pope/UQ AT QH.E4 A/F 1995 Atherton Tablelands, Qld L. Pope/UQ AT QH.E5 F May 5, 96 Atherton Tablelands, Qld L. Pope/UQ AT QH.E6 F Ma~. 96 Atherton Tablelands, Qld L. Pope/UQ AR QH.M1 F Dec. 14,32 Atherton Tablelands, Qld MoV- DTC 275 AR QH.M2 M Dec. 14,32 Atherton Tablelands, Qld MoV-DTC276 AR QH.M3 F Dec. 14,32 Atherton Tablelands, Qld MoV-DTC277 AR QH.M4 F Dec. 17,32 lower Archer River, Qld MoV-DTC 282 AR QH.M5 M Dec. 20,32 lower Archer River, Qld MoV-DTC284 AR QH.M6 M Dec.20,32 lower Archer River, Qld MoV-DTC285 AR QH.M7 M Dec. 20,32 lower Archer River, Qld MoV-DTC286 AR QH.M8 F Dec. 20,32 lower Archer River, Qld MoV-DTC287 AR QH.M9 M Dec. 20,32 lower Archer River, Qld MoV-DTC 288 AR QH.Mu1 M Dec. 14,32 Archer River estuary, Qld MoV-DTC 278 AR QH.Mu2 M Dec. 14,32 Archer River estuary, Qld MoV-DTC 279 AR QH.Mu3 M Dec. 14,32 Archer River estuary, Qld MoV-DTC 280 AR QH.Mu4 F Jan 27, 33 lower Archer River, Qld MoV-DTC 289 AR QH.Mu5 F June 13, 33 lower Archer River, Qld MoV-DTC 290 AR QH.Mu6 F Sept. 33 lower Archer River, Qld MoV-DTC294 Dasr_urus g_eoffroii BF WG.E1 F Aug. 8, 94 Batalling State Forest, WA G. Haii!PZ {BSF} BF WG.E2 F Au~. 8, 94 BSF G. Haii/PZ BF WG.E3 M Aug. 8, 94 BSF G. Haii/PZ BF WG.E4 M Au~. 8, 94 BSF G. Haii/PZ BF WG.E5 M Aug. 8, 94 BSF G. Haii/PZ BF WG.E6 F Aug. 8, 94 BSF G. Haii/PZ BF WG.E8 Aug. 8, 94 BSF G. Haii/PZ BF WG.E9 Aug. 8, 94 BSF G. Haii/PZ BF WG.E10 Au~. 8, 94 BSF G. Haii/PZ BF WG.E11 Aug. 8, 94 BSF G. Haii/PZ BF WG.E12 G. Haii/PZ BF WG.E13 G. Haii/PZ BF WG.E14 G. Hall/PZ BF WG.E15 G. Haii/PZ BF WG.B1 M Sept. 92 Captive born G. Haii/PZ BF WG.B2 F Aug./Sept. 92 Captive born G. Haii/PZ

202 Species/ 10#2 Age/ Date Location collected Collector/ 4 Pop1 Sex3 collected lnstitution BF WG.B3 G. Haii/PZ BF WG.B4 M Sept. 92 Captive born G. Haii/PZ BF WG.B5 G. Haii/PZ BF WG.B6 F Sept. 92 Captive born G. Haii/PZ BF WG.B7 M Sept. 92 Captive born G. Haii/PZ BF WG.B8 F Sept. 92 Captive born G. Haii/PZ BF WG.B9 M Sept. 92 Captive born G. Haii/PZ BF WG.B10 M Sept. 92 Captive born G. Haii/PZ BF WG.B11 F Sept. 92 Captive born G. Haii/PZ BF WG.B12 F Sept. 92 Captive born G. Haii/PZ BF WG.B13 F Sept. 92 Captive born G. Haii/PZ BF WG.B14 F Sept. 92 Captive born G. Haii/PZ BF WG.B15 M Sept. 92 Captive born G. Haii/PZ BF WG.B16 M Sept. 92 Captive born G. Haii/PZ BF WG.B17 G. Haii!PZ BF WG.B18 F Sept. 92 Captive born G. Haii/PZ BF WG.B19 M Sept. 92 Captive born G. Haii/PZ BF WG.B20 M Sept. 92 Captive born G. Haii/PZ BF WG.B21 M Sept./Oct. 92 Captive born G. Haii/PZ WG.E7 NM Aug. 17, 94 Lake Magenta, WA G. Haii/PZ S: 33°28' E: 119°07' PE WG.M1 1917 Perth Area WAM 311 S: 31 56 00 E: 115 56 00 PE WG.M2 1928 Perth Area WAM 1046 S: 32 00 00 E: 115 59 00 PE WG.M3 1929 Perth Area WAM 1106 S: 31 57 30 E: 115 50 40 PE WG.M4 1930 Perth Area WAM 1234 S: 32 22 00 E: 115 58 00 PE WG.M5 1930 Perth Area WAM 1291 S: 32 22 00 E: 115 58 00 PE WG.M6 1930 Perth Area WAM 1326 S: 32 22 00 E: 115 58 00 PE WG.M7 1930 Perth Area WAM 1337 S: 31 57 00 E: 115 5100 PE WG.M8 1933 Perth Area WAM 1736 S: 32 22 00 E: 115 58 00 PE WG.M9 1934 Perth Area WAM 1842 S: 31 54 00 E: 116 08 00 PE WG.M10 1934 Perth Area WAM 1863 S: 32 03 00 E: 116 00 00 PE WG.M11 1934 Perth Area WAM 1865 S: 31 54 00 E: 116 08 00 PE WG.M12 1934 Perth Area WAM 1866 S: 31 54 00 E: 116 08 00 PE WG.M13 1936 Perth Area WAM 2062 S: 31 54 00 E: 116 08 00 PE WG.M14 1936 Perth Area WAM 2063 S: 3210 00 E: 116 03 00 PE WG.M15 1940 Perth Area WAM 2444 S: 31 54 00 E: 116 08 00 PE WG.M16 1955 Perth Area WAM 3075 S: 32 01 00 E: 116 07 00

203 -

2 Species/ 10# Age/ Date Location collected Collector/ 4 Pop1 Sex3 collected lnstitution PE WG.M17 Jun. 12, 58 Perth Area WAM 3359 S: 32 20 00 E: 116 07 00 PE WG.M18 May 28, 59 Perth Area WAM 4464 S: 31 46 00 E: 116 23 00 PE WG.M19 Apr. 1, 62 Perth Area WAM 4969 S: 3217 45 E: 116 15 00 PE WG.M20 Jun. 7,65 Perth Area WAM 6582 S: 32 37 00 E: 116 28 00 PE WG.M21 Nov. 27,66 Perth Area WAM 7574 S: 32 07 00 E: 116 14 00 PE WG.M22 1968 Perth Area WAM 13715 S: 31 47 00 E: 116 11 00 PE WG.M23 1900 Perth Area WAM 16034 S: 31 59 00 E: 115 52 00 PE WG.M24 1909 Perth Area WAM 16039 S: 31 57 00 E: 115 51 00 Das'(_urus se_artacus NGS.D1 May 8, 96 New Guinea M. Westerman/LU Das'(_urus alboe_unctatus NGA.L1 AIM Oct. 10, 85 Bobole, Southern Highlands S. Donnellan/SAM Province, New Guinea J49 S: 6° 15' E: 142° 47' Sarcophilus harrisii TS.D1 Tasmania D. Colgan/AM

1 Population codes follow that of Table 6. 2 ID numbers are as follows: for most samples the first letter indicates the state, second letter indicates the species, third letter indicates the tissue type. For samples from New Guinea and the Northern Territory the first two letters indicate the country or state, the third letter indicates the sepcies and the fourth letter indicates the tissue type. Thus, NM.E1 is a New South Wales maculatus ear tissue sample and NTH. L 15 is a Northern Territory hallucatus liver sample. 3 Age and Sex codes are as follows: F = female, M = male, J =juvenile, A = adult, PY = pouch ¥oung. Institutional abbreviations: AM = Australian Museum, ANU = Australian National Universtiy, DCNR = Department of Conservation and Natural Resources, FWP = Featherdale Wildlife Park, LU = LaTrobe University, MoV = Museum of Victoria, MoV-DTC = Museum of Victoria­ Donald Thompson Collection, PZ =Perth Zoo, QVM =Queen Victoria Museum, SAM= South Australian Museum, TZ = Taronga Zoo, TWP = Trowunna Wildlife Park, UQ = University of Queensland, UTAS = University of Tasmania, WAM =Western Australian Museum.

204 Appendix 3. Microsatellite genotypes of all quolls analysed in these studies. Populations are labelled as in Table 6, ID codes follow Appendix 2. SPECIES/ POPULATION QUOLLID 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 Dasyurus maculatus co NM.E1 96/100 157/157 105/107 116/116 80/90 197/197 co NM.E2 96/98 157/157 103/105 114/114 82/84 199/199 co NM.E3 98/98 157/157 99/103 116/120 86/90 187/193 co NM.E4 157/157 103/105 84/86 195/197 co NM.E5 98/98 157/157 103/109 116/120 82/82 193/195 co NM.E6 98/98 157/159 103/107 118/120 80/90 195/195 co NM.E7 98/98 157/157 103/105 116/116 80/82 193/199 co NM.L1 96/98 159/165 103/105 116/116 84/88 195/197 co NM.L 11 98/98 157/159 105/107 120/120 80/86 187/197 BG NM.E9 98/98 159/161 103/105 114/116 80/82 197/197 BG NM.E10 98/98 159/161 103/103 116/116 84/90 193/197 BG NM.E11 98/98 159/159 103/105 114/120 80/88 197/197 BG NM.E12 96/98 159/159 105/107 116/120 80/90 189/197 BG NM.E13 96/98 157/159 103/105 114/116 80/90 189/197 BG NM.E14 98/98 157/159 103/103 120/120 80/82 193/193 BG NM.E15 98/98 159/159 107/107 116/116 80/84 193/199 BG NM.E17 96/96 157/157 103/103 120/120 88/90 193/197 BG NM.E18 96/98 159/165 103/103 120/124 86/86 193/197 BG NM.E19 96/96 157/159 103/107 116/120 90/90 189/197 BG NM.E20 96/96 157/159 114/120 90/90 BG NM.E24 96/98 157/159 105/107 116/120 82/86 187/197 BG NM.E33 96/98 159/159 107/109 116/128 86/90 193/197 BG NM.E36 96/98 165/165 103/107 116/128 80/82 193/195 BG NM.E37 96/98 159/159 105/109 116/116 90/90 193/197 BG NM.E38 98/100 157/165 103/105 114/116 82/84 193/197 BG NM.L3 98/98 157/159 103/103 120/120 80/82 193/193 BG NM.T6 98/98 159/159 103/105 116/116 84/90 197/197 BG NM.T7 98/98 159/159 103/105 116/116 84/90 197/197 BG NM.T8 96/98 159/159 105/105 116/116 84/90 197/197 BG NM.T9 98/98 159/159 103/105 116/116 84/90 197/197 BG NM.T10 98/98 159/159 103/105 116/116 84/90 197/197 BG NM.T11 96/98 159/159 105/105 116/116 84/90 197/197 CH NM.E16 98/98 159/159 103/109 116/128 80/86 191/193 CH NM.E21 98/100 149/157 103/107 134/134 80/84 195/199 CH NM.E22 98/98 159/161 103/103 116/120 86/86 187/197 CH NM.E23 98/98 159/159 103/107 116/116 82/82 187/193 CH NM.E25 98/98 157/159 107/107 116/134 82/82 187/199 CH NM.E26 96/98 157/165 103/103 116/120 86/90 193/201 CH NM.E28 98/100 157/157 103/107 116/120 82/84 193/199 CH NM.E29 96/98 157/159 103/107 116/120 80/80 193/193 CH NM.E30 98/98 159/165 105/109 116/116 82/82 193/197 CH NM.E32 96/98 157/159 103/103 116/120 82/86 193/195 CH NM.E40 96/98 157/157 103/103 116/120 82/86 187/193 CH NM.E41 96/96 157/157 105/107 116/120 80/90 193/199 CH NM.L4 98/98 157/157 103/103 116/120 82/90 187/197 CH NM.T1 98/98 157/157 103/107 116/120 80/86 193/201

205 SPECIES/ POPULATION QUOLLID 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 CH NM.T2 96/96 157/159 103/103 116/120 80/90 193/199 CH NM.T3 96/96 157/157 103/107 120/120 80/90 193/201 CH NM.T4 96/98 157/159 103/103 116/120 80/86 193/201 CH NM.T5 96/96 157/157 105/107 116/120 80/90 193/199 BT NM.E8 96/98 159/161 103/103 116/116 86/86 195/199 BT NM.E27 (96/98) 157/165 103/107 82/86 197/199 BT NM.E31 78/88 BT NM.E35 98/98 159/165 103/105 86/86 187/197 BT NM.E39 98/98 157/157 103/107 82/88 195/199 BT NM.L2 96/96 159/159 103/105 116/120 84/86 199/199 BT NM.L7 96/98 157/165 103/105 80/82 195/197 BT NM.L8 98/100 157/157 103/105 116/116 84/90 197/197 BT NM.L 10 98/98 161/165 103/107 116/116 76/84 199/199 BT NM.L 13 98/98 157/157 103/103 86/90 187/193 BS NM.E42 96/96 159/159 103/103 86/88 187/187 BS NM.E43 98/98 157/157 103/103 116/120 84/88 187/193 BS NM.E44 96/98 159/159 103/103 116/118 86/88 187/193 BS NM.E45 98/98 159/159 101/103 114/114 84/88 187/187 BS NM.E46 96/98 157/157 103/103 116/116 88/88 187/187 BS NM.E47 98/98 159/159 103/103 116/116 84/88 187/193 Gl NM.B1 96/96 157/159 103/107 116/116 80/90 197/201 Gl NM.B2 96/96 159/165 101/103 82/82 Gl NM.B3 96/96 159/159 103/103 118/118 82/82 199/199 Gl NM.Mu 1 96/96 159/165 103/103 116/118 82/82 199/201 Gl NM.E34 96/96 105/105 116/118 80/82 201/201 TT TM.E1 96/98 163/163 103/103 108/116 86/86 197/199 TT TM.E2 98/98 161/163 103/105 116/120 82/86 195/199 TT TM.E3 98/98 159/161 103/103 116/120 84/84 195/197 TT TM.E4 98/98 159/161 105/105 118/120 84/86 199/203 TT TM.E5 98/98 159/159 105/105 114/116 84/84 195/199 TT TM.E6 98/98 161/161 103/103 84/84 195/195 TT TM.E7 98/98 161/161 101/103 118/120 86/86 195/199 TT TM.E8 98/98 157/163 103/103 120/120 84/86 195/195 TT TM.E9 98/98 153/159 103/105 120/120 84/84 195/197 TT TM.E10 96/98 161/163 105/105 118/128 82/84 195/197 TT TM.E11 98/98 161/163 105/105 118/118 84/86 195/197 WY TM.M1 82/84 WY TM.M2 155/155 99/99 WY TM.M3 96/96 84/84 WY TM.M4 153/155 99/99 WY TM.M5 99/99 96/96 WY TM.M6 90/96 99/99 132/132 84/96 WY TM.M7 82/82 WY TM.M8 82/84 WY TM.M9 99/99 82/84 WY TM.M10 96/96 WY TM.M11 84/84 193/197 WY TM.M12 153/155 82/84 WY TM.M13 WY TM.M14 94/94 99/99 82/84

206 SPECIES/ POPULATION QUOLLID 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 WY TM.M15 80/82 WY TM.M16 99/99 187/187 WY TM.M17 153/157 WY TM.M18 132/134 74/84 WY TM.M19 132/132 WY TM.M20 84/86 WY TM.M21 99/99 78/84 WY TM.M22 130/132 82/84 WY TM.M23 82/84 WY TM.M24 132/132 82/84 WY TM.M25 86/86 WY TM.M26 99/99 82/86 WY TM.M27 99/99 82/84 WY TM.M28 WY TM.M29 126/126 86/86 WY TM.M30 WY TM.M31 96/96 WY TM.M32 84/84 WY TM.M33 82/86 76/84 WY TM.M34 82/84 76/76 WY TM.M35 82/84 99/99 MW QM.E1 100/100 157/161 105/105 114/120 80/80 199/199 MW QM.E2 98/98 157/157 105/105 114/114 80/90 199/199 MW QM.E3 92/100 157/157 103/105 120/120 90/92 193/193 MW QM.E4 100/100 157/161 103/105 114/114 90/92 193/193 MW QM.E5 98/98 157/157 105/105 114/114 80/90 193/193 MW QM.E6 98/98 157/161 105/105 114/120 80/90 193/193 MW QM.E7 92/98 157/157 103/105 114/120 80/80 193/195 MW QM.E8 98/100 157/157 105/105 120/120 92/92 193/199 MW QM.E9 98/98 157/157 105/105 114/114 80/90 193/193 MW QM.E10 98/98 157/157 103/105 114/120 80/92 193/193 MW QM.E11 100/100 157/161 103/105 120/120 80/92 193/199 MW QM.L1 157/161 105/105 80/90 193/199 SB VM.L2 100/100 159/159 103/103 118/124 86/88 197/197 SB VM.E3 96/98 159/165 101/103 116/118 76/86 187/197 SB VM.E4 96/98 159/159 103/103 116/116 76/86 187/197 Dasyurus viverrinus TE TV.E1 96/96 149/151 105/105 138/138 96/98 213/215 TE TV.E2 92/96 149/149 105/105 138/140 96/96 191/213 TE TV.E3 92/92 149/151 103/107 134/134 96/96 213/217 TE TV.E4 92/96 149/151 105/105 138/138 98/98 215/215 TE TV.E5 92/96 149/149 103/103 134/142 96/96 191/193 TE TV.E6 96/96 149/149 103/105 144/144 193/215 TE TV.E7 96/96 149/151 103/105 134/142 96/96 189/213 TE TV.E8 92/92 149/151 103/105 140/140 98/98 215/215 TE TV.E9 96/96 149/149 105/105 138/140 96/96 213/213 TE TV.E10 92/92 149/149 105/105 138/140 96/96 191/213 TE TV.E11 92/96 151/151 103/103 140/140 96/96 213/215 TE TV.E12 98/98 149/151 103/105 140/140 92/96 215/215 TE TV.E13 92/96 149/149 103/105 134/138 96/96 191/215 TE TV.E14 96/98 149/149 103/105 134/138 96/96 215/217

207 SPECIES/ POPULATION QUOLLID 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 VB TV.E15 96/96 105/105 140/142 94/98 215/215 VB TV.E16 92/96 149/151 103/103 138/140 94/98 191/191 VB TV.E17 92/92 151/151 103/105 140/140 94/98 191/215 VB TV.E18 92/92 149/149 105/105 138/138 94/98 215/215 VB TV.E19 96/96 149/151 105/105 142/142 98/98 191/191 VB TV.E20 92/96 149/151 103/103 134/140 96/98 191/215 VB TV.E21 92/92 149/151 103/105 138/140 94/98 191/215 VB TV.E22 92/92 151/151 105/105 140/140 94/98 191/215 VB TV.E23 92/96 151/151 103/105 138/138 98/98 191/191 VB TV.E24 96/96 147/149 105/105 134/140 92/96 191/215 VB TV.E25 92/92 149/151 103/105 134/134 94/96 191/215 VB TV.E26 92/92 149/149 103/103 138/138 96/98 191/191 VB TV.E27 92/92 149/149 105/105 134/134 92/94 215/215 VB TV.E28 92/92 149/149 103/105 138/140 94/98 215/215 VB TV.E29 96/96 149/149 103/103 138/138 96/98 191/191 VB TV.E30 92/96 149/151 103/105 134/138 94/94 215/215 VB TV.E31 92/96 149/151 103/103 138/140 92/98 191/191 VB TV.E32 96/96 149/151 105/105 126/130 94/98 215/215 VB TV.E33 92/96 149/149 103/103 126/126 94/98 191/191 VB TV.E34 96/96 151/151 105/105 126/126 96/98 213/215 VB TV.E35 92/96 149/151 105/105 126/128 94/96 191/191 GL TV.M1 92/92 94/96 213/213 GL TV.M2 92/96 149/149 103/105 138/142 94/98 211/213 GL TV.M3 149/149 140/142 90/96 GL TV.M4 96/96 149/151 134/138 96/96 193/213 GL TV.M5 92/96 149/149 103/105 130/134 96/96 211/213 GL TV.M6 92/96 149/149 103/105 136/140 96/1 02 213/213 GL TV.M7 96/96 149/151 140/142 94/96 213/213 GL TV.M8 94/96 149/151 142/142 90/96 213/213 GL TV.M9 96/96 149/149 140/142 94/94 213/213 GL TV.M10 92/96 149/149 101/103 136/140 90/102 191/213 GL TV.M11 92/96 149/149 140/142 90/96 211/213 GL TV.M12 96/96 149/149 103/105 130/134 96/96 211/213 GL TV.M13 96/96 149/149 138/142 96/96 211/213 GL TV.M14 92/96 149/149 136/140 94/96 213/213 GL TV.M15 96/96 149/149 105/105 138/142 96/96 213/213 GL TV.M16 96/96 149/149 103/105 138/142 90/96 213/213 GL TV.M17 96/96 149/149 103/105 130/134 96/96 213/213 GL TV.M18 96/96 149/149 103/105 140/142 94/96 213/215 GL TV.M19 96/96 149/149 103/105 130/134 94/96 211/213 GL TV.M20 96/96 149/149 105/105 130/134 96/96 211/213 GL TV.M21 96/96 149/149 105/105 136/140 94/94 213/213 GL TV.M22 92/96 149/149 103/103 130/134 90/94 213/213 GL TV.M23 96/96 149/151 103/105 130/134 96/96 211/213 GL TV.M24 92/96 149/149 103/105 140/144 96/96 211/213 GL TV.M25 149/149 101/103 130/134 94/96 211/213 GL TV.M26 92/96 149/149 130/134 92/102 211/211 GL TV.M27 149/151 105/105 211/213 GL TV.M28 96/96 149/149 103/105 138/142 96/96 189/211 GL TV.M29 92/96 149/149 101/103 130/134 92/96 213/213 GL TV.M30 92/92 149/149 103/105 138/142 96/98 213/213

208 SPECIES/ POPULATION QUOLLID 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 GL TV.M31 96/96 149/149 103/103 138/142 94/94 213/213 GL TV.M32 96/96 149/149 101/103 130/134 94/96 211/213 GL TV.M33 96/96 149/149 103/105 130/134 94/96 213/215 GL TV.M34 92/96 101/103 130/134 96/96 211/213 GL TV.M35 92/92 149/149 105/105 130/134 96/102 211/213 GL TV.M36 92/96 149/149 103/105 140/142 94/96 213/213 GL TV.M37 92/96 149/149 103/105 138/142 96/96 213/213 GL TV.M38 92/96 103/105 130/134 96/96 213/213 GL TV.M39 96/96 149/151 101/103 130/134 96/96 213/215 GL TV.M40 94/96 149/151 105/105 138/142 96/96 211/213 GL TV.M41 96/96 105/105 130/134 96/96 211/211 GL TV.M42 96/96 149/149 101/103 90/96 213/213 GL TV.M43 96/96 149/149 103/105 130/134 90/96 213/213 GL TV.M44 96/96 149/149 101/103 138/142 94/96 213/213 GL TV.M45 92/96 149/149 103/105 136/140 90/96 213/213 GL TV.M46 96/96 149/149 105/105 138/140 90/96 211/213 GL TV.M47 96/96 149/149 101/103 138/140 96/96 213/213 GL TV.M48 96/96 101/103 90/94 213/213 GL TV.M49 96/96 149/149 105/105 130/134 96/96 213/213 GL TV.M50 92/96 149/149 105/105 138/142 94/94 211/211 GL TV.M51 149/151 103/105 130/134 96/96 213/213 GL TV.M52 149/149 103/105 GL TV.M53 161/161? 96/96 GL TV.M54 103/103 GL TV.M55 96/96 149/149 101/103 130/134 94/96 211/213 GL TV.. M56 96/96 149/149 101/103 130/134 90/96 211/211 GL TV.M57 92/96 149/149 101/103 130/134 94/96 209/211 GL TV.M58 96/96 149/149 101/103 138/142 96/96 211/211 ST W.M1 96/96 101/103 ST W.M2 103/103 ST W.M3 ST W.M4 132/132 ST W.M5 ST W.M6 ST W.M? ST W.M8 105/105 118/118 ST W.M9 ST W.M10 ST W.M11 94/96 ST W.M12 96/96 105/105 ST W.M13 94/94 ST W.M14 ST W.M15 ST W.M16 157/159 ST W.M17 96/96 99/99 ST W.M18 96/96 105/105 ST W.M19 94/96 ST W.M20 94/94 ST W.M21 96/96 146/146 NE NV.M1 98/100 101/105 120/120 195/195 NE NV.M2 92/96 103/103 1181120 84/84

209 --

SPECIES/ POPULATION QUOLLID 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 NE NV.M3 103/103 118/120 84/84 NE NV.M4 94/96 103/105 118/120 NE NV.M5 88/88 163/163 103/103 118/118 76/84 NE NV.M6 101/101 120/120 78/88 NE NV.M7 98/98 101/101 110/120 100/100 NE NV.M8 101/103 116/120 88/96 NE NV.M9 169/169 101/103 114/114 82/92 NE NV.M10 95/101 120/120 NE NV.M11 149/167 105/105 120/120 70/70 NE NV.M12 92/96 143/157 99/107 118/120 92/96 NE NV.M13 96/96 101/103 118/120 70/80 NE NV.M14 99/103 120/120 NE NV.M15 96/98 101/101 114/120 80/82 NE NV.M16 99/103 118/120 84/84 NE NV.M17 103/105 116/116 NE NV.M18 96/96 147/147 101/101 88/96 NE NV.M19 95/103 118/120 NE NV.M20 96/96 101/105 120/120 84/98 NE NV.M21 98/98 101/103 120/120 78/82 NE NV.M22 98/98 103/103 118/120 90/90 NE NV.M23 96/98 149/149 118/120 94/96 NE NV.M24 92/92 143/143 99/103 116/118 98/98 NE NV.M25 92/92 151/153 107/107 191/199 Dasyurus hallucatus KA NTH.L 1 82/82 147/151 91/91 128/138 90/90 179/179 KA NTH.L2 82/82 149/163 91/93 124/124 70/88 179/181 KA NTH.L3 82/82 153/165 91/101 116/116 88/94 179/179 KA NTH.L4 82/82 151/163 91/113 128/148 90/92 179/179 KA NTH.L5 82/84 143/155 91/111 128/128 84/92 179/181 KA NTH.L6 82/86 147/149 91/97 128/146 90/92 179/179 KA NTH.L7 86/86 153/153 97/119 128/128 88/96 179/181 KA NTH.L8 82/84 143/155 91/111 126/126 84/92 179/181 KA NTH.L9 82/82 147/149 91/99 128/128 90/92 179/179 KA NTH.L 10 82/82 147/151 91/91 140/148 70/90 179/181 KA NTH.L 11 80/82 151/151 91/103 126/136 86/92 179/179 KA NTH.L 12 82/86 145/153 91/95 128/144 70/90 179/181 KA NTHL.13 82/82 147/147 91/91 124/138 70/70 179/183 KA NTH.L 14 82/82 147/147 91/91 136/142 96/98 179/181 KA NTH.L 15 82/84 155/161 107/119 128/142 88/88 179/179 KA NTH.E1 82/86 147/151 93/99 90/92 179/181 KA NTH.E2 82/82 153/159 91/95 118/118 92/92 179/181 KA NTH.E3 82/84 147/161 91/91 126/136 88/94 179/179 KA NTH.E4 82/82 147/147 91/91 118/128 90/90 179/179 KA NTH.E5 82/86 149/151 91/91 138/138 70/92 179/181 KA NTH.E6 82/82 147/149 91/97 126/132 70/92 179/181 KA NTH.E7 82/86 145/163 91/91 128/128 100/100 179/179 KA NTH.E8 82/88 149/149 91/91 126/132 90/92 179/181 KA NTH.E9 82/82 145/145 91/97 134/138 70/88 179/179 KA NTH.E10 82/86 145/155 91/101 136/142 90/100 179/181 KA NTH.E11 82/84 149/149 91/93 88/92 AT QH.E1 84/84 151/153 95/117 130/138 82/86 179/179

210 SPECIES/ POPULATION QUOLL ID 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 AT QH.E2 84/84 147/149 93/111 132/132 88/88 179/179 AT QH.E3 84/84 147/149 113/113 126/132 88/92 179/179 AT QH.E4 84/84 147/147 93/109 132/142 86/88 179/179 AT QH.E5 84/84 147/149 105/111 116/132 88/88 179/179 AT QH.E6 84/84 149/149 93/109 116/132 94/94 179/179 AR QH.M1 82/96 101/103 82/84 AR QH.M2 82/96 145/145 95/111 130/130 84/84 AR QH.M3 96/96 90/94 AR QH.M4 AR QH.M5 147/149 82/84 AR QH.M6 103/103 84/84 AR QH.M7 149/149 130/134 80/84 AR QH.M8 82/84 AR QH.M9 101/103 80/96 AR QH.Mu1 AR QH.Mu2 AR QH.Mu3 96/98 80/84 AR QH.Mu4 AR QH.Mu5 AR QH.Mu6 Dasyurus geoffroii BF WG.E1 90/98 157/157 107/107 132/138 84/84 203/205 BF WG.E2 90/102 147/147 111/141 11 0/132 82/11 0 207/209 BF WG.E3 90/100 147/147 111/113 132/134 82/1 02 209/209 BF WG.E4 98/98 157/157 107/137 132/132 78/84 205/207 BF WG.E5 90/104 157/161 111/111 11 0/138 90/90 213/217 BF WG.E6 90/98 157/161 131/143 11 0/138 82/90 207/213 BF WG.E8 94/102 153/153 110/132 88/110 207/207 BF WG.E9 94/98 151/157 107/135 132/132 78/78 203/209 BF WG.E10 104/104 157/161 141/141 110/138 82/82 205/205 BF WG.E11 94/102 153/155 111/133 132/148 88/90 205/205 BF WG.E12 104/106 157/161 110/138 82/90 205/211 BF WG.E13 94/96 155/157 135/143 132/132 80/84 201/209 BF WG.E14 94/102 153/157 111/135 132/134 78/11 0 205/205 BF WG.E15 104/104 161/161 131/143 134/144 82/90 205/211 BF WG.B1 90/102 157/161 111/131 11 0/134 82/90 205/209 BF WG.B2 104/104 130/134 84/90 207/207 BF WG.B3 90/98 157/157 109/143 110/138 82/90 205/215 BF WG.B4 90/106 157/161 131/137 110/144 90/90 209/211 BF WG.B5 90/98 157/161 131/143 11 0/138 82/90 205/211 BF WG.B6 90/98 157/161 131/137 138/138 90/90 205/211 BF WG.B7 104/106 157/161 111/143 11 0/138 90/90 207/209 BF WG.B8 98/98 151/157 107/137 132/132 78/84 203/205 BF WG.B9 104/104 110/110 82/82 203/209 BF WG.B10 90/106 161/161 131/139 134/138 78/78 209/211 BF WG.B11 96/98 151/155 107/137 126/132 78/84 201/209 BF WG.B12 96/106 109/143 118/120 82/84 BF WG.B13 90/98 161/161 111/137 110/144 82/82 207/209 BF WG.B14 98/104 161/161 131/143 110/138 82/90 205/211 BF WG.B15 94/98 151/157 137/145 132/132 80/84 199/203 BF WG.B16 120/120 84/90

211 SPECIES/ POPULATION QUOLL 10 1.3 3.1.2 3.3.1 3.3.2 4.4.2 4.4.10 BF WG.B17 90/106 157/161 131/143 11 0/144 90/90 205/211 BF WG.B18 104/106 157/161 111/137 110/138 207/209 BF WG.B19 104/106 157/161 131/143 11 0/138 82/90 205/211 BF WG.B20 98/98 157/157 107/137 132/132 78/78 203/205 BF WG.B21 104/106 161/165 131/137 110/144 90/90 209/211 LAKE MAGENTA WG.E7 98/98 153/153 135/137 126/138 82/90 201/201 PE WG.M1 101/103 120/120 PE WG.M2 101/103 PE WG.M3 130/134 PE WG.M4 102/102 101/103 82/86 PE WG.M5 103/105 PE WG.M6 109/111 211/213 PE WG.M7 PE WG.M8 92/92 PE WG.M9 94/96 82/88 PE WG.M10 94/94 101/103 136/136 78/84 PE WG.M11 84/88 PE WG.M12 90/92 PE WG.M13 96/96 101/103 PE WG.M14 207/207 PE WG.M15 98/98 139/139 84/84 PE WG.M16 96/96 151/157 137/139 205/205 PE WG.M17 98/98 155/161 113/131 138/148 205/207 PE WG.M18 94/106 153/157 119/127 90/98 187/205 PE WG.M19 100/104 153/153 113/113 134/136 201/201 PE WG.M20 100/102 155/157 137/141 110/144 209/211 PE WG.M21 96/100 147/157 138/146 209/213 PE WG.M22 96/104 147/157 113/113 134/138 201/205 PE WG.M23 96/96 147/157 82/90 PE WG.M24 102/110 145/151 142/142 84/92 Dasyurus spartacus NEW GUINEA NGS.1 100/102 153/153 105/109 126/138 92/92 201/201

212 11111111112222222222333333333344444444445555555555666666666677777 Taxon/Node 12345678901234567890123456789012345678901234567890123456789012345678901234 1 GCACAA------TTTTAATTTTAATCACAAAA-TTCAATTTATATGCTAT 2 GCACAA------TTTTAATCTTAATCACAAAA-TTCAATTTATATGCTAT 3 GCACAA------TTTTAATTTTAATCACAAAA-TTTAATTTGTATGCTAT 4 GCACAA------TTTTAATTTCAATCACAAAA-TTTAATTTGTATGCTAT 5 GCACAA------TTTTAATTTTAATCACAAAA-TTTAATTTGTATGCTAT 6 GCACAA------TTTTAATTTTAATCACAAAA-TTTAATTTGTATGCTAT 7 GCACAA------TTTTAATTTTAATCACAAAA-TTCAATTTATATGCTAT 8 ???????????????????????????????????????????????????????A-TTCAATTAATATGCTAT 9 GCACAA------TTTTAATTTTAATCACAAAA-TTCAATTTATATGCTAT 10 GCACAA------TTTTAATTTTAATCACAAAA-TTCAATTTATATGCTAT 11 GCACAA------TTTTAATTTTAATCACAAAA-TTCAATTTATATGCTAT 12 GCACAA------TTTTAATTTTAATCACAAAA-TTCAATTTATATGCTAT 13 ??????????????????????????????????TTTTTTTTATTTTATCGAAAAA-TTTAATTTATATGCTCT 14 ?????????????????????????????????????????????????????????????????????????? 15 GCACAAA------TTTATTTTTTTTTTATTTTATCAAAAAA-TTTAATTTATATGCTAT 16 GCCCAAA------TTTATTTTTTTTTTATTTTATCAAAAAA-TTTAATTTATATGCTAT 17 ?????????????????????????????????????????????????????????????????????????? 18 ?????????????????????????????????????????????????????????????????????????? 19 GCACAAA------TTATTTTTTTTTATTTTTTTTTACAAAAATTCAATTTATATGCTAT 20 GCACAAA------TTATTTTTTTTTTATTTTTTTTACAAAAATTCAATTTATATGCTAT 21 GCCCAAA------TTTTTTTTTTATTTTTTTTTTTACAAAAATTCAATTTATATGCTAT 22 GCACAAACTTATTTCTTTATTTTACACAATTTTTTTTTCTTTATTTTATTACAAAA-TTTAATTTATATGCTAT 23 ------GTAAACACTTCCATTATTTTATCTTTTTTTTT-CATTTT-ATTT-AAAATTTTAATTTATATACTAT 24 ------GTAGATGCTTCCATTATTTTATTTTTTTTT---AATTTTTATTT-AAAATTTTAATTTATATACTAT 25 ------GTAAACACTTCCATTATTTTATCTTTTTTTTTTCATTTTTATTT-AAAATTTTAATTTATATACTAT 26 ------GTAAACACTTCCATTATTTTATCTTTTTTTT--CATTTTTATTT-AAAATTTTAATTTATATACTAT 27 ------GTAAACACTTCCATTATTTTATCTTTTTTTT---ATTTT-ATTT-AAAATTTTAATTTATATACTAT 28 ------GTAAACACTTCCATTATTTTATCTTTTTTTTT-CATTTTTATTT-AAAATTTTAATTTATATACTAT 29 ------GTAAACACTTCCATTATTTTATCTTTTTTT---CATTTTTATTT-AAAATTTTAATTTATATACTAT 30 ------GCTTAAATTTATTTTTTTTTTTAATTTTAATTTATATACTAT 31 ------CCATCTGTTTCTGCAAAAATCAATTTATATACTAT 32 ------ACAACAAGGTAATGCAATTTAATATTCCACTATTACTTTAATTTTATATACTAT Appendix 4. The mtDNA control region alignment. Sequences are numbered as in Table 3.

213 1111111111111111111111111111111111111111111111111 77777888888888899999999990000000000111111111122222222223333333333444444444 Taxon/Node 56789012345678901234567890123456789012345678901234567890123456789012345678 1 CTCAGTATTAAATTTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 2 CTCAGTATTAAATTTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 3 CTCAGTATTAAAATTTTT-----A------CAAAA---TTTTTCA------CAAAATTTTT---AAT 4 CTCAGTATTAAAATTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 5 CTCAGTATTAAAATTTTT-----A------CAAAA---TTTTTCA------CAAAATTTTT---AAT 6 CTCAGTATTAAAATTTTT-----A------CAAAA---TTTTTCA------CAAAATTTTT---AAT 7 CTCAGTATTAAATTTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 8 CTCAGTATTAAATTTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 9 CTCAGTATTAAATTTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 10 CTCAGTATTAAATTTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 11 CTCAGTATTAAATTTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 12 CTCAGTATTAAATTTTTT-----A------CAAAA---ATTTTCA------CAAAATTTTT---AAT 13 CTCAGTATTAAATTTTTTTT-CTA------CAAA-TTTTTTTTTA------CAAA-TTTCT---AAC 14 ??????ATTAAATTTTTTTTT-TA------CAAA--TTTTTTTTA------CAAA-TTTCT---AAN 15 CTCAGTATTAAATTTTTTTT-CTA------CAAAATTTTTTTTTA------CAAA-TTTCT---AAC 16 CTCAGTATTAAATTTTTTTTTCTA------CAAA--TTTTTTTTA------CAAA-TTTCT---AAC 17 ?????????????????????????????????AAATTTTTTTTTA------CAAA-TTTCT---AAC 18 ?????????????????????????????????????????????????????????????????????????? 19 CCCAGTATTAAATTTTT------A------CAAAAA-TTTTT------AATTTTT---AAT 20 CCCAGTATTAAATTTTT------A------CAAAAA-TTTTT------AATTTTT---AAT 21 CCCAGTATTAAATTTTT------A------CAAAAAATTTTT------AATTTTT---AAT 22 CTCAGTATTAAATTTTTTTTT--A------CAA------AATTTTT---AAT 23 CTCAGTATTAAATTTTTT------CAAAA------TTTTTTT---ACT 24 CTCAGTATTAAATTTTTT------CAAAA------TTTTTT----ATT 25 CTCAGTATTAAATTTTTT------CAAAA------TTTTTTT---ACT 26 CTCAGTATTAAATTTTTT------CAAAA------TTTTTTT---ACT 27 CTCAGTATTAAATTTTTT------CAAAA------TTTTTTT---ACT 28 CTCAGTATTAAATTTTTT------CAAAA------TTTTTTT---ACT 29 CTCAGTATTAAATTTTTT------CAAAA------TTTTTTT---ACT 30 CCCAGTATTAATTTTTTTTT---A------CAA--TTTTTTATAAAATCAATTTAAACAAATATACGCA-CAT 31 CCCAGTATTAAATTTTTTAATTTATTTTAATAAAAAAAAAACAACGTAACACGAGTATTGTAGTTCGTAC-TAT 32 GTCAGTATTAAATTCATTAT------CAAATTTTAATTGCCTAGACATACAATTATATAAGTACATTAA Appendix 4. Continued

214 11111111111111111111111111111111111111111111111111122222222222222222222222 45555555555666666666677777777778888888888999999999900000000001111111111222 Taxon/Node 90123456789012345678901234567890123456789012345678901234567890123456789012 1 TCAAACATAC--ATT---ACAT-TTTACATGTATAC-A-GTGTATAAAAATTTA-TTTATGTATATAGAGCATA 2 TCAAACATAC--ATT---ATAT-TTTACATGTATAT-A-ATGTACAAAAATTTA-TTTATGTATATAGAGCATA 3 TCAAACATAT--ATT---ACAT-TTTATATGTATAT-A-ATGCACAAAAATTTA-TTTATGTATATAGAGCATA 4 TCAAACATAC--ATT---ACAT-TTTACATATATAT-A-ATGCACAAAAATTTA-TTTATGTATATAGAGCATA 5 TCAAACATAT--ATT---ACAT-TTTACATGTATAT-A-ATGCACAAAAATTTA-TTTATGTATATAGAGCATA 6 TCAAACATAC--ATT---ACAT-TTTACATGTATAT-A-ATGCACAAAAATTTA-TTTATGTATATAGAGCATA 7 TCAAACATAC--ATT---ACAT-TTTACATGTATAC-A-GTGTACAAAAATTTA-TTTATGTATATAGAGCATA 8 TCAAACATAC--ATT---ACAT-TTTACATGTATAC-A-GTGTACAAAAATTTA-TTTATGTDTATAGAGCATA 9 TCAAACATAC--ATT---ACAT-TTTACATGTATAC-A-GTGTATAAAAATTTA-TTTATGTATATAGAGCATA 10 TCAAACATAC--ATT---ACAT-TTTACATGTATAC-A-GTGTACAAAAATTTA-TTTATGTATATAGAGCATA 11 TCAAACATAC--ATT---ACAT-TTTACATGTATAC-A-GTGTACAAAAATTTA-TTTATGTATATAGAGCATA 12 TCAAACATAC--ATT---ATAT-TTTACATGTATAT-A-ATGTACAAAAATTTA-TTTATGTATATAGAGCATA 13 TTAAACAACTT-ATT---ATATATTTACATATATATTA-ATGTGTAAATATATA-ATTATGTATATAGAGCATA 14 TTAAACAACTT-ATT---ATATATTTACATATATATTA-ATGTGTAAATATATA-ATTATGTATATAGAGCATA 15 TTAAACAACTT-ATT---ATATATTTACATATATATTA-ATGTGTAAATATATA-ATTATGTATATAGAGCATA 16 TTAAACAACTT-ATT---ATATATTTACATATATATTA-ATGTGTAAATATATA-ATTATGTATATAGAGCATA 17 TTAAACAACTT-ATT---ATATATTTACATATATATTA-ATGTGTAA-TATATA-ATTATGTATATAGAGCAAA 18 ??AAAAAAATT-MWT---A-ATATATATATATATATTA-ATGTGTAAATATATA-ATTATGTATATAGAGCATA 19 TCAAATAAATATATT---ATATATTTATATATACATTA-ATGTATAAATATATA-ATTATGTATATAGAGCATA 20 TCAAATAAATCTATT---ATATATTTATATATACATCT-ATGTATAAATATATA-ATTATGTATATAGAGCATA 21 TCAAGTAAATATATT---ATATATTTATATATACATCA-ATGTATAAATATATA-ATTATGTATATAGAGCATA 22 TCAAA-AAATATACA---CTCTACTATTACATACATTTTATGTATAATCATATA-ATTATGTATATAGAGCATA 23 TTAAATAAAT--AT----ATCTTTATACATTTATATTA-ATGTATAAAGATATA-TTTATGTATATAGAGCATA 24 TTAAATAAAT--AT----ATCTTTATACATTTATATTA-ATGTATAAAGATATA-TTTATGTATATAGAGCATA 25 TTAAATAAAT--AT----ATCTTTATACATTTATATTA-ATGTATAAAGATATA-TTTATGTATATAGAGCATA 26 TTAAATAAAT--AT----ATCCTTATACATTTATATTA-ATGTATAAAGATATA-TTTATGTATATAGAGCATA 27 TTAAATAAAT--AT----ATCTTTATACATTTATATTA-ATGTATAAAGATATA-TTTATGTATATAGAGCATA 28 TTAAATAAAT--AT----ATCCTTATACATTTATATTA-ATGTATAAAGATATA-TTTATGTATATAGAGCATA 29 TTAAATAAAT--AT----ATCCTTATACATTTATATTA-ATGTATAAAGATATA-TTTATGTATATAGAGCATA 30 TCATATAGATTCATCC--ATATTCTGATATACAAATCA-ATATAATATGATATA-TTTATGTATATAGAGCATA 31 ACAAACAACTACACCACACTCAATGTAAGCTGGCATCA-ATGTAATTATCCATATAGTATGTATATAGAGCATT 32 TGCCACAAATACATA---CTAT-TCCCCACATACATAATAATATAACAAATACATAATATGTATATATTACATT Appendix 4. Continued

215 22222222222222222222222222222222222222222222222222222222222222222222222222 22222223333333333444444444455555555556666666666777777777788888888889999999 Taxon/Node 34567890123456789012345678901234567890123456789012345678901234567890123456 1 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 2 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATG-T 3 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 4 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 5 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 6 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 7 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 8 CATTTATATTCCTCTATAATATAAG-CMTGTD-YATATT-AATCATATATTACTAAATACATTAATATAATA-T 9 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AACCATATATTACTAAATACATTAATATAATA-T 10 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 11 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAACATAATA-T 12 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATG-T 13 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 14 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 15 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 16 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 17 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 18 CATTTATATTCCTTTATMATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAAWA-T 19 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 20 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTGATATAATA-T 21 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 22 CATTTATATTCCTCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAATATAATA-T 23 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAAACTAATA-T 24 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAAACTATTA-T 25 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAAACTAATA-T 26 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAGATACATTAAATTAATA-T 27 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAAACTAATA-T 28 CATTTATATTCCCCTATAATATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAAACTAATA-T 29 CATTTATATTCCCCTATAGTATAAG-CATGTA-CATATT-AATCATATATTACTAAATACATTAAATTAATA-T 30 CATTTATATTCCTCTATCATATAATTAATGTA-CATAAT-AATCATATATTACTAAATACATTAATGTAATA-T 31 AATTTAATGTACACTATCATATAAT-AGTATA-CATATT-AATCATATATTACTAAATACATTAATTTATTAAT 32 CAATTATTTACCACTAGCATATAAGCTATATAGTACATTCAATCAATTAT-ACTA-AGACATAAAAGATAAA-C Appendix 4. Continued

216 22233333333333333333333333333333333333333333333333333333333333333333333333 99900000000001111111111222222222233333333334444444444555555555566666666667 Taxon/Node 78901234567890123456789012345678901234567890123456789012345678901234567890 1 ATTACTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAATTCATAATAGTACATAATACATAATA 2 ATTATTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAACTCATAATAGTACATAATACATAATA 3 ATTGTTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAATTCATAATAGTACATAATACATAATA 4 ATTGTTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAATTCATAATAGTACATAATACATAATA 5 ATTGTTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAATTCATAATAGTACATAATACATAATA 6 ATTGTTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAATTCATAATAGTACATAATACATAATA 7 ATTACTATCAT--TTGAACAGTACATTCATATCTTAAACTAATG-CAACTCATAATAGTACATAATACATAATA 8 ATTACTGTCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAACTCATAATAGTACATAATACATAATA 9 ATTACTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAATTCATAATAGTACATAATACATAATA 10 ATTACTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAACTCATAATAGTACATAATACATAATA 11 ATTACTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAACTCATAATAGTACATAATACATAATA 12 ATTGTTATCAT--TTGAACAGTACATTCATATCTTAAACTAATA-TAACTCATAATAGTACATAATACATAATA 13 ATTGCTATAAT--TTAATTAATACATTCATATCTTATACTAATA-TCATTCATAATAGTACATAATACATAATA 14 ATTGCTATAAT--TTAATTAATACATTCATATCTTATACTAATA-TCATTCATAATAGTACATAATACATAATA 15 ATTGCTATAAT--TTAATTAATACATTCATATCTTATACTAATA-TCATTCATAATAGTACATAATACATAATA 16 ATTGCTATAAT--TTAATTAATACATTCATATCTTATACTAATA-TCATTCATAATAGTACATAATACATAATA 17 ATTGCTATAAT--TTAATTAATACATTCATATCTTATACTAATA-TCATTCATAATAGTACATAATACATAATA 18 ATTGCTAAAMA--TTAATAAATACATKCATATCTTAAACTAATA-TAATTCATAATAGTACATAATACATAATA 19 ATTGCTATAAAA-TTAATTAATACATTCTTATCTTAAATTAATAATAACTCATAATAGTACATAATACATAATA 20 ATTGCTATAAAA-TTAATTAATACATTCTTATCTTAAGTTAATAATAACTCATAATAGTACATAATACATAATA 21 ATTGCTATAAA--TTAATTAATACATTCTTATCTTAAATTAATAATTACTCATAATAGTACATAATACATAATA 22 ATTGCTATAAT--TTAATCAGTACATTCATATCTTATACTAATA-TAACTCATAATAGTACATAATACATAATA 23 ATTACTAATAA--TTAATAGATACATTCATATCTTAACCTAAGTATAAATCATAATAGTACATAATACATAATA 24 ATTGCTAAGAA--TTAACATATACATTCATATCTTAACCTAAATATAAATCATAATAGTACATAATACATAATA 25 ATTACTAATAA--TTAATAGATACATTCATATCTTAACCTAAGTATAAATCATAATAGTACATAATACATAATA 26 ATTACTAATAA--TTAATAAACACATTCATATCTTAACCTAAGTATAAATCATAATAGTACATAATACATAATA 27 ATTACTAATAA--TTAATAAATACATTCATATCTTAACCTAAGTATAAATCATAATAGTACATAATACATAATA 28 ATTACTAATAA--TTAATAAACACATTCATATCTTAACCTAAGTATAAATCATAATAGTACATAATACATAATA 29 ATTACTAATAA--TTAATAAACACATTCATATCTTAACCTAAGTATAAATCATAATAGTACATAATACATAATA 30 ATTACTAATAA--TTAATTAATACATTCATATCTCTTACTAATA-TAACTCATAATAGTACATTATACATAATA 31 AGTACTAATAATGTTATTATATACATTCATATCTTTAATACATA-TATCTAATAATAGTACATAATACATATAA 32 ATTACTATAAA--TTAAATAAAGCATTCATATCTCAAATACATAATA-CTCATA-TATAACATAAAACATTATA Appendix 4. Continued

217 33333333333333333333333333333444444444444444444444444444444444444444444444 77777777788888888889999999999000000000011111111112222222222333333333344444 Taxon/Node 12345678901234567890123456789012345678901234567890123456789012345678901234 1 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTATATCCATGGAACGA-GATCAACAAAG 2 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACGA-GATCAACAAAG 3 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACAA-GATCAACTAAG 4 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACAA-GATCAACTAAG 5 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACAA-GATCAACTAAG 6 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACAA-GATCAACTAAG 7 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACGA-GATCAATAAAG 8 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTATATCCATGGAACGA-GATCAACAAAG 9 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTATATCCATGGAACGA-GATCAACAAAG 10 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACGA-GATCAACAAAG 11 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACGA-GATCAACAAAG 12 TGTATATATT-ACATAATACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACGA-GATCAACAAAG 13 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGATCGA-GATCAGTAAAG 14 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGATCGA-GATCAGTAAAG 15 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGATCGA-GATCAGTAAAG 16 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGATCGA-GATCAGTAAAG 17 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGATCGA-GATCAGTAAAG 18 TGTATATATT-ACATAAGACATAWAATG-TTGGCGTACATAGACATTAAATCCATGGATCAA-GATCAAYAAAG 19 TGTATATATT-ACATAAGACATATAATG-TTGACGTACATAGACATTAAATCCATGGAACGA-GATCAATAAAG 20 TGTATATATT-ACATAAGACATATAATG-TTGACGTACATAGACATTAAATCCATGGAACGA-GATCAATAAAG 21 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACGA-GATCAATAAAG 22 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAAATCCATGGAACAA-GATCAATAAAG 23 TGTATAAATT-ACATAATACATATAATG-TTGGAGTACATAGACATTAATTCCATGGAACGG-GATCAG-AAAG 24 TGTATAAGTT-ACATAATACATATAATG-TTGGAGTACATAGACATTAATTCCATGGAACGG-GATCAT-AAAG 25 TGTATAAATT-ACATAATACATATAATG-TTGGAGTACATAGACATTAATTCCATGGAACGG-GATCAA-AAAG 26 TGTATAAGTT-ACATAATACATATAATG-TTGGAGTACATAGACATTAATTCCATGGAACGG-GATCAA-AAAG 27 TGTATAAGTT-ACATAATACATATAATG-TTGGAGTACATAGACATTAATTCCATGGAACGG-GATCAG-AAAG 28 TGTATAAGTT-ACATAATACATATAATG-TTGGAGTACATAGACATTAATTCCATGGAACGG-GATCAA-AAAG 29 TGTATAAGTT-ACATAATACATATAATG-TTGGAGTACATAGACATTAATTCCATGGAACGG-GATCAA-AAAG 30 TGTATATATT-ACATAAGACATATAATG-TTGGCGTACATAGACATTAACTCCATGAATCAA-GACCACTAAAA 31 TGTATATATT-ACATAATACATATAATG-TTGGCGTACATAGACATATCATCCATGATTCAT-GATCAAAC-AC 32 TGTATAACTTTACATA-TACATTAATAGCTTAATCCACATAAACATAAAATCAATGAAAGAAAGATCAGACAAC Appendix 4. Continued

218 44444444444444444444444444444444444444444444444444444445555555555555555555 44444555555555566666666667777777777888888888899999999990000000000111111111 Taxon/Node 56789012345678901234567890123456789012345678901234567890123456789012345678 1 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 2 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 3 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 4 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTCCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 5 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 6 CTAATTCCTCTAGCAGACCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 7 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCCCCCCTCACGAGAGATCAGCAACCCGCCA 8 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 9 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 10 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 11 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 12 CTAATTCCTCTAGCAGATCACAACCATTAGGATT-ACCTTCATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 13 CTAATTCCTCTAGCAGATCATCACCATTAGGAAT-ACCTTTATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 14 CTAATTCCTCTAGCAGATCATCACCATTAGGAAT-ACCTTTATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 15 CTAATTCCTCTAGCAKATCATCACCATTAGGAAT-ACCTTAATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 16 CTAATTCCTCTAGCATATCATCACCATTAGGAAT-ACCTTAATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 17 CTAATTCCTTTAGCAGATCATCACCATTAGGAAT-ACCTTAATCCTCCCCTCACGAGAGATCAGCAACCCGCCA 18 CTAATTCCTCTAGCAGATCATCACCATTAGGAAT-ACCTTTATCCTCCACTCACGAGAGATCAGCAACCCGCCA 19 CTAATTCCTCTAGCAGATCATCACCATAGGGAAT-ACCTTTATCCTCCCCTCACGAGAGATCAGCAACCCGTCA 20 CTAATTCCTCTAGCAGATCATCACCATAGGGAAT-ACCTTTATCCTCCCCTCACGAGAGATCAGCAACCCGTCA 21 CTAATTCCTCTAGCAGATCATCACCATTGGGAAT-ACCTTTATCCTCCCCTCACGAGAGATCAGCAACCCGTCA 22 CTAATTCCTCTAGCAGATCATCACCATTAGGAAT-ACCTTTATCCTCCCCTCACGAGAGATCAGCAACCCGTCA 23 CTTTATCCTCAAGCAGATCACTACCATTAGGAAT-ACCTTTATCCCCAACTCACGAGAGATCAGCAACCCGCCA 24 CTTTATCCTCAAGCAGATCACAACCATTAGGGGT-ACCTTTATCCCCAACTCACGAGAGATCAGCAACCCGCCA 25 CTTTATCCTCAAGCAGATCACTACCATTAGGAAT-ACCTTTATCCCCAACTCACGAGAGATCAGCAACCCGCCA 26 CTTTATCCTCAAGCAGATCATAACCATTAGGAAT-ACCTTTATCCCCAACTCACGAGAGATCAGCAACCCGCCA 27 CTTTATCCTCAAGCAGATCACTACCATTAGGAAT-ACCTTTATCCCCAACTCACGAGAGATCAGCAACCCGCCA 28 CTTTATCCTCAAGCAGATCACAACCATTAGGAAT-ACCTTTATCCCCAACTCACGAGAGATCAGCAACCCGCCA 29 CTTTATCCTCAAGCAGATCACAACCATTAGGAAT-ACCTTTATCCCCAACTCACGAGAGATCAGCAACCCGCCA 30 CTACTTCCTCTAGCAGATCATCACCATTACGGAT-ACCTTAATCCCCAACTCACGAGAGATCAGCAACCCGCCA 31 CTTTATCCTCTAGCAGATCATTACCTTTAGGATTTACCTTACTCCACACCTCACGAGAGATCAGCAACCCGCCA 32 CTCATTACTCTAGCATATCCCTACCNTTAGTAAT-CCCTTTATCCCCAACTCACGAGAAACCACCATCCCGCCA Appendix 4. Continued

219 5555555555555555555555555555555555 1222222222233333333334444444444555 Taxon/Node 9012345678901234567890123456789012 1 TTCATAGACACAATATCCTTCAGAGCAAGCCCAT 2 TTCATAGACACAACATCCTTCAGAGCAAGCCCAT 3 TTCATAGACACAACATCCTTCAGAGCAAGCCCAT 4 TTCATAGACACAACATCCTTCAGAGCAAGCCCAT 5 TTCATAGACACAACATCCTTCAGAGCAAGCCCAT 6 TTCATAGACACAACATCCTTCAGAGCAAGCCCAT 7 TTCATAGACACAACATCCTTCAGAGCAAGCCCAT 8 CTCATAGACACAACATCCTTCAGAGCAAGCCCAT 9 TTCATAGACACAATATCCTTCAGAGCAAGCCCAT 10 TTCACAGACACAACATCCTTCAGAGCAAGCCCAT 11 TTCATAGACACAACATCCTTCAGAGCAAGCCCAT 12 TTCATAGACACAACATCCTTCAGAGCAAGCCCAT 13 TTYGTAGACACWATATCCTTCAGAGCAAGCCCAT 14 TTYGTAGACACWATATCCTTCAGAGCAAGCCCAT 15 TTCATAGACACWATATCCTTCAGAGCAAGCCCAT 16 TTCATAGACACWATATCCTTCAGAGCAAGCCCAT 17 TTCGTAGACACAATATCCTTCAGAGCAAGCCCAT 18 TCTGWAGACACTAAATCCTTCAGAGCAAGCCCAT 19 TTCACAGTCACTACATCCTTCAGAGCAGGCC??? 20 TTCACAGACACTACATCCTTCAGAGCAGGCCCAT 21 TTCATAGACACTACATCCTTCAGAGCAGGCCCAT 22 TCCATAGACACTACATCCTTCAGAGCAGGCCCAT 23 TTCATAGACACAACATCCTTCAGAGCAGGCCCAT 24 TCCATAGACACAACATCCTTCAGAGCAGGCCCAT 25 TTCATAGACACAACATCCATCAGAACAGGCCCAT 26 TTCATAGACACAACATCCTTCAGAGCAGGCCCAT 27 TTCATAGACACAACATCCTTCAGAGCAGGCCCAT 28 TTCATAGACACAACATCCTTCAGAGCAGGCCCAT 29 TCCATAGACACAACATCCTTCAGAGCAGGCCCAT 30 TCTGAAGAT-CAATATCCTTCAGAGCAAGCCCAT 31 TTCATAGACACGAACTCCTTCAGAGCAGGCCCAT 32 TCCAAAGGCTTAACATCCTTCAGAGCAAGCCCAT Appendix 4. Continued

220 15 34 PE BF AT KA AR TE VB GL ST NE TT WY SB BS BT BG CH CO GI 11 5 8 12 16 10 6 3 6 11 15 9 51 21 14 26 4 6 MW 0.000 0.0000.000 0.000 0.0000.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.091 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.250 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.167 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 0.083 0.000 0.0000.545 0.000 0.000 0.000 0.000 0.000 0.188 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.364 0.250 0.067 0.000 0.000 0.750 0.000 0.000 0.000 0.019 0.083 0.000 0.406 0.000 0.000 0.0000.000 0.667 0.000 0.000 0.000 0.000 0.000 0.063 0.250 0.000 0.712 0.000 0.000 0.563 0.000 0.0000.000 0.000 0.000 0.000 0.083 0.333 0.000 0.250 0.000 0.000 0.700 0.000 0.096 0.000 0.167 0.000 0.000 0.031 0.000 0.3330.000 0.000 0.200 0.000 0.667 0.000 0.000 0.000 0.000 0.000 0.000 0.050 0.154 0.250 0.000 0.000 0.000 0.000 0.000 0.3330.000 1.000 0.000 0.033 0.000 0.000 0.000 0.091 0.000 0.000 0.225 0.000 0.019 0.000 0.000 0.000 0.000 0.333 0.000 0.333 0.000 0.367 0.000 0.548 0.000 0.000 0.909 0.000 0.000 0.000 0.020 0.000 0.000 0.000 0.667 0.000 0.393 0.000 0.000 0.300 0.000 0.000 0.000 0.000 0.000 0.000 0.755 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.033 0.452 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.500 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.191 0.000 0.000 0.000 0.067 0.000 0.107 0.000 0.000 0.000 0.000 0.625 0.000 0.000 0.000 0.000 0.000 0.133 0.000 0.000 0.000 0.000 0.000 0.125 0.000 0.088 0.000 0.000 0.300 0.000 0.000 0.000 0.000 0.000 0.000 0.044 0.000 0.000 0.133 0.000 0.000 0.000 0.000 0.000 0.235 0.000 0.000 0.100 0.000 0.000 0.000 0.000 0.015 0.000 0.133 0.000 0.000 0.000 0.074 0.000 0.067 0.000 0.000 0.221 0.033 0.000 0.132 0.033 0.000 POPULATION (N) Appendix 5. Allele frequency distributions for microsatellite loci in four species (20 populations) of quolls. Population codes are as Table 6. (N) = sample size. LOCUS 1.3 80 82 84 86 88 90 92 94 96 98 100 102 104 106 110

221 9 31 PE BF AT 3 6 KA AR TE VB 52 20 14 26 GL ST 8 1 NE 11 TT WY 6 3 4 BS SB BT BG 12 16 10 CH CO 4 9 GI 12 MW 0.000 0.0000.000 0.000 0.0000.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.042 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.792 0.000 0.000 0.000 0.000 0.000 0.188 0.000 0.125 0.0000.000 0.000 0.000 0.000 0.000 0.0000.778 0.000 0.000 0.625 0.000 0.0000.208 0.000 0.000 0.0000.458 0.000 0.0000.167 0.000 0.125 0.000 0.000 0.0000.000 0.0000.250 0.000 0.0000.375 0.000 0.000 0.000 0.188 0.000 0.000 0.500 0.0000.000 0.0000.563 0.375 0.0000.042 0.000 0.000 0.038 0.333 0.063 0.000 0.250 0.200 0.0450.000 0.0250.063 0.500 0.9040.000 0.000 0.000 0.0000.056 0.096 0.667 0.063 0.000 0.000 0.100 0.0000.000 0.5500.000 0.000 0.125 0.0770.083 0.333 0.833 0.000 0.250 0.000 0.000 0.714 0.000 0.000 0.000 0.045 0.4250.125 0.000 0.000 0.000 0.167 0.000 0.000 0.173 0.000 0.000 0.286 0.063 0.200 0.056 0.227 0.000 0.417 0.000 0.000 0.500 0.000 0.135 0.500 0.000 0.000 0.000 0.000 0.000 0.167 0.409 0.000 0.417 0.000 0.000 0.000 0.167 0.500 0.096 0.000 0.065 0.000 0.000 0.000 0.000 0.273 0.083 0.000 0.000 0.000 0.000 0.000 0.000 0.077 0.000 0.000 0.125 0.000 0.111 0.000 0.000 0.083 0.000 0.019 0.000 0.000 0.000 0.000 0.065 0.000 0.000 0.167 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.019 0.065 0.063 0.000 0.111 0.000 0.000 0.000 0.000 0.000 0.000 0.038 0.048 0.125 0.000 0.333 0.000 0.000 0.000 0.000 0.000 0.058 0.403 0.000 0.000 0.000 0.000 0.000 0.000 0.019 0.000 0.000 0.056 0.000 0.000 0.000 0.000 0.339 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.016 0.000 0.000 0.000 0.000 0.000 POPULATION (N) Appendix 5. Continued 143 145 147 149 151 153 155 157 159 161 163 165 167 169 LOCUS 3.1.2

222 14 30 PE BF AT 4 6 KA AR TE VB 47 21 14 26 GL 6 ST NE TT 11 11 24 WY 6 3 BS SB BT BG 12 15 10 CH CO 5 9 GI 12 MW 0.000 0.0000.000 0.000 0.0000.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.056 0.000 0.100 0.000 0.0000.208 0.000 0.000 0.000 0.000 0.000 0.000 0.600 0.0000.792 0.000 0.000 0.000 0.000 0.389 0.000 0.000 0.200 0.000 0.0000.000 0.000 0.000 0.542 0.000 0.0000.333 0.000 0.000 0.100 0.000 0.0000.000 0.467 0.000 0.0000.083 0.000 0.0000.167 0.083 0.042 0.000 0.650 0.000 0.0000.2330.000 1.000 0.0000.292 0.167 0.0000.056 0.917 0.000 0.000 0.200 0.000 0.0000.000 0.0000.233 0.000 0.0000.083 0.833 0.000 0.615 0.000 0.000 0.083 0.000 0.000 0.150 0.0450.000 0.0000.067 0.000 0.000 0.000 0.000 0.000 0.1670.000 0.058 0.000 0.313 0.000 0.000 0.000 0.500 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0830.000 0.038 0.000 0.000 0.375 0.000 0.000 0.000 0.455 0.0000.000 0.250 0.000 0.1490.000 0.125 0.000 0.077 0.2500.000 0.000 0.000 0.125 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.083 0.000 0.000 0.4260.000 0.000 0.500 0.038 0.000 0.000 0.000 0.063 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.4290.000 0.000 0.426 0.000 0.000 0.000 0.000 0.038 0.000 0.000 0.000 0.393 0.000 0.000 0.000 0.000 0.5710.000 0.000 0.000 0.000 0.000 0.250 0.000 0.000 0.019 0.000 0.000 0.571 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.500 0.000 0.000 0.000 0.000 0.000 0.000 0.036 0.000 0.000 0.179 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.019 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.214 0.0000.000 0.083 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.036 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.038 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.167 0.000 0.000 0.000 0.125 0.000 0.019 0.000 0.117 0.000 0.000 0.000 0.000 0.036 0.000 0.000 0.000 0.167 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.033 0.000 0.000 0.000 0.000 0.036 0.000 0.167 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.038 0.000 0.167 0.000 0.000 0.179 0.000 0.000 0.000 0.083 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.017 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.036 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.036 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.036 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.183 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.017 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.050 0.000 0.071 0.000 0.000 0.000 0.000 0.167 0.000 0.107 0.000 0.000 0.000 0.017 0.036 0.000 0.000 0.050 0.000 0.000 0.167 0.000 0.017 POPULATION (N) Appendix 5. Continued 91 93 95 97 99 101 103 105 107 109 111 113 117 119 127 131 133 135 137 139 141 143 145 LOCUS 3.3.1

223 9 35 PE BF AT 2 6 KA AR TE VB 51 21 14 24 GL 3 ST NE 10 23 TT WY BS SB 5 5 3 6 BT BG 12 16 CH CO 4 8 GI 11 MW 0.000 0.0000.000 0.000 0.0000.545 0.000 0.000 0.0000.000 0.000 0.000 0.125 0.500 0.000 0.0000.000 0.000 0.500 0.000 0.000 0.5000.455 0.156 0.542 0.000 0.063 0.000 0.000 0.0000.000 0.438 0.000 0.000 0.000 0.313 0.200 0.000 0.800 0.0500.000 0.000 0.000 0.292 0.000 0.000 0.600 0.000 0.000 0.000 0.000 0.3130.000 0.000 0.000 0.500 0.000 0.100 0.022 0.200 0.000 0.0500.000 0.031 0.000 0.000 0.333 0.000 0.000 0.100 0.065 0.000 0.000 0.2000.000 0.000 0.000 0.000 0.042 0.000 0.000 0.000 0.087 0.000 0.000 0.000 0.250 0.0000.000 0.063 0.000 0.000 0.167 0.000 0.000 0.000 0.000 0.261 0.000 0.000 0.400 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.3330.000 0.000 0.565 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.167 0.000 0.0000.125 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.042 0.000 0.000 0.000 0.000 0.000 0.056 0.000 0.050 0.0000.000 0.000 0.000 0.083 0.000 0.000 0.000 0.000 0.063 0.271 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.167 0.667 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.1430.000 0.000 0.083 0.000 0.000 0.000 0.000 0.063 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.0240.000 0.000 0.2250.000 0.000 0.000 0.3330.000 0.125 0.000 0.014 0.000 0.000 0.000 0.111 0.000 0.000 0.0240.000 0.000 0.000 0.000 0.000 0.000 0.000 0.292 0.043 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.083 0.000 0.2350.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.1670.000 0.000 0.000 0.750 0.049 0.000 0.000 0.042 0.000 0.014 0.000 0.214 0.000 0.000 0.000 0.083 0.000 0.000 0.147 0.000 0.000 0.000 0.021 0.000 0.000 0.000 0.000 0.056 0.000 0.310 0.500 0.000 0.137 0.250 0.000 0.000 0.083 0.014 0.000 0.321 0.000 0.000 0.262 0.000 0.000 0.196 0.000 0.000 0.104 0.271 0.000 0.321 0.167 0.000 0.071 0.000 0.010 0.000 0.333 0.021 0.086 0.000 0.071 0.167 0.000 0.083 0.000 0.000 0.000 0.063 0.000 0.071 0.167 0.000 0.000 0.000 0.000 0.021 0.200 0.000 0.000 0.000 0.083 0.000 0.021 0.000 0.000 0.111 0.000 0.000 0.042 0.000 0.056 0.000 0.000 0.071 0.056 0.000 0.000 0.056 0.014 POPULATION (N) Appendix 5. Continued 108 110 114 116 118 120 124 126 128 130 132 134 136 138 140 142 144 146 148 LOCUS 3.3.2

224 9 34 PE BF AT KA AR TE VB GL 0.000 0.0000.000 0.000 0.0000.000 0.154 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.167 0.000 0.000 0.000 0.056 0.000 0.000 0.000 0.167 0.147 0.000 0.038 0.000 0.000 0.083 0.000 0.500 0.029 0.109 0.019 0.167 0.000 0.000 0.000 0.000 0.250 0.018 0.154 0.278 0.000 0.167 0.071 0.000 0.147 0.209 0.231 0.056 0.038 0.500 0.333 0.056 0.000 0.609 0.250 0.111 0.000 0.000 0.167 0.000 0.029 0.018 0.038 0.167 0.769 0.083 0.429 0.056 0.338 0.000 0.038 0.111 0.192 0.167 0.000 0.056 0.000 0.036 0.019 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.058 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.056 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.015 0.000 0.044 ------ST NE TT WY SB BS BT BG CH CO GI 12 5 9 12 16 11 6 3 26 11 18 0 55 21 13 26 9 6 MW 0.000 0.0000.000 0.000 0.0000.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.0000.458 0.000 0.000 0.000 0.000 0.200 0.000 0.0000.000 0.000 0.000 0.222 0.045 0.000 0.700 0.000 0.0000.000 0.208 0.000 0.000 0.222 0.045 0.019 0.000 0.219 0.3330.000 0.083 0.375 0.000 0.167 0.000 0.045 0.058 0.000 0.156 0.0000.000 0.083 0.000 0.000 0.000 0.167 0.182 0.019 0.000 0.094 0.0000.292 0.028 0.250 0.000 0.000 0.056 0.136 0.019 0.100 0.125 0.0000.250 0.056 0.000 0.250 0.000 0.167 0.318 0.250 0.000 0.063 0.0000.000 0.056 0.083 0.167 0.000 0.091 0.091 0.385 0.000 0.344 0.5000.000 0.000 0.083 0.583 0.545 0.000 0.136 0.115 0.000 0.000 0.1670.000 0.222 0.000 0.000 0.364 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.083 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.056 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.056 0.000 0.000 0.000 0.000 0.000 0.135 0.000 0.000 0.028 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.111 0.000 0.000 0.000 0.000 0.000 0.083 0.000 0.000 0.000 0.000 0.056 0.000 0.000 0.000 0.000 0.000 POPULATION (N) Appendix 5. Continued 70 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 110 LOCUS 4.4.2

225 9 33 PE BF AT 1.000 0.0000.000 0.000 0.0000.000 0.000 0.0000.000 0.000 0.0560.000 0.000 0.0000.000 0.000 0.0000.000 0.000 0.0000.000 0.000 0.0000.000 0.000 0.0000.000 0.000 0.0000.000 0.015 0.1670.000 0.030 0.0000.000 0.091 0.2780.000 0.288 0.1670.000 0.152 0.1110.000 0.212 0.1110.000 0.152 0.1110.000 0.030 0.0000.000 0.015 0.000 0.015 ------0 6 KA AR TE VB 54 21 14 25 GL 0.000 0.0000.000 0.000 0.0000.000 0.720 0.000 0.0000.000 0.260 0.000 0.0000.009 0.020 0.000 0.0000.009 0.000 0.036 0.5240.009 0.000 0.143 0.0000.000 0.000 0.071 0.0000.000 0.000 0.000 0.0000.000 0.000 0.000 0.0000.000 0.000 0.000 0.0000.000 0.000 0.000 0.0000.000 0.000 0.000 0.0000.000 0.000 0.000 0.0000.009 0.000 0.000 0.0000.269 0.000 0.000 0.0000.667 0.000 0.000 0.0240.028 0.000 0.286 0.4520.000 0.000 0.393 0.000 0.000 0.071 0.000 ------ST 2 0 NE 11 TT WY 6 3 2 BS SB BT BG 12 15 10 CH CO 4 9 GI 12 MW 0.000 0.0000.000 0.000 0.0000.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.111 0.000 0.000 0.0000.000 0.000 0.167 0.000 0.000 0.000 0.000 0.0000.667 0.033 0.000 0.000 0.000 0.000 0.000 0.000 0.150 0.0000.042 0.100 0.000 0.042 0.000 0.167 0.750 0.000 0.000 0.000 0.000 0.000 0.417 0.333 0.278 0.000 0.000 0.125 0.000 0.0000.292 0.333 0.500 0.083 0.000 0.278 0.000 0.000 0.375 0.050 0.000 0.033 0.000 0.083 0.000 0.167 0.250 0.000 0.500 0.150 0.000 0.467 0.000 0.167 0.000 0.000 0.000 0.000 0.300 0.000 0.033 0.250 0.042 0.000 0.000 0.000 0.250 0.000 0.350 0.000 0.000 0.000 0.000 0.667 0.000 0.000 0.000 0.000 0.5000.000 0.000 0.250 0.000 0.000 0.000 0.000 0.500 0.000 0.000 0.2270.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.2270.000 0.000 0.000 0.000 0.000 0.000 0.000 0.250 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.0450.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 POPULATION (N) Appendix 5. Continued LOCUS 4.4.10 179 181 183 187 189 191 193 195 197 199 201 203 205 207 209 211 213 215 217

226 Appendix 6. Allele frequencies for 6 microsatellite loci I 9 populations of tiger quolls. Populations are as in Table 11; allelic designations are listed at left. Number of individuals analysed for each population/locus and all alleles unique to a population are shown in bold.

Population Locus COP BGH CSF BAD MAN BAR TAS QLD SUG Q1.3 8 16 12 6 5 10 11 11 3 G .000 .000 .000 .000 .000 .000 .000 .091 .000 I .188 .406 .250 .333 1. 000 .250 . 091 .000 .333 J .750 .563 .667 . 667 .000 .700 .909 .545 .333 K .063 .031 .083 .000 .000 .050 .000 .364 .333 Q3.1.2 9 16 12 6 4 10 11 12 3 D .000 .000 .042 .000 .000 .000 .000 .000 .000 F .000 .000 .000 .000 .000 .000 .045 .000 .000 H .778 .250 .458 .333 .125 .500 .045 .792 .000 I .167 .563 .375 .667 . 625 .200 .227 .000 .833 J .000 .063 .042 .000 .000 .100 .409 .208 .000 K .000 .000 .000 .000 .000 .000 .273 .000 .000 L .056 .125 .083 .000 .250 .200 .000 .000 .167 Q3.3.1 9 15 12 6 5 10 11 12 3 E .056 .000 .000 .000 .000 .000 .000 .000 .000 F .000 .000 .000 .083 .100 .000 .045 .000 .167 G .389 .467 .542 .917 • 600 .650 .500 .208 .833 H .333 .233 .083 .000 .200 .200 .455 .792 .000 I .167 .233 .292 .000 .100 .150 .000 .000 .000 J .056 .067 .083 .000 .000 .000 .000 .000 .000 Q3.3.2 7 16 12 5 4 5 10 11 3 c .143 .156 .000 .200 .000 .000 .050 .545 .000 D .429 .438 .542 .600 .500 .800 .200 .000 .500 E .071 .000 .000 .100 .500 .000 .250 .000 .333 F .357 .313 .292 .100 .000 .200 .400 .455 .000 G .000 .031 .000 .000 .000 .000 .000 .000 .167 I .000 .063 .042 .000 .000 .000 .050 .000 .000 L .000 .000 .125 .000 .000 .000 .000 .000 .000 Q4.4.2 9 16 12 6 5 11 11 12 3 c .000 .000 .000 .000 .000 .045 .000 .000 .333 D .000 .000 .000 .000 .000 .045 .000 .000 .000 E .222 .219 .208 .000 .200 .045 .000 .458 .000 F .222 .156 .375 .000 .700 .182 .091 .000 .000 G .167 .094 .083 .250 .000 .136 .545 .000 .000 H .167 .125 .250 .167 .000 .318 . 364 .000 .500 I .056 .063 .000 .583 .000 .091 .000 .000 .167 J .167 .344 .083 .000 .100 .136 .000 .292 .000 K .000 .000 .ooo .000 .000 .000 .000 .250 .000 Q4.4.10 9 15 12 6 4 10 11 12 3 D .111 .033 .167 .750 .000 .150 .000 .000 .333 F .000 .000 .042 .000 .000 .000 .000 .000 .000 G .167 .333 .417 .250 .000 .050 .000 . 667 .000 H .278 .033 .083 .000 .000 .150 .500 .042 .000 I .278 . 467 .083 .000 .125 .300 .227 .000 .667 J .167 .033 .167 .000 .375 .350 .227 .292 .000 K .000 .000 .042 .000 .500 .000 .000 .000 .000 L .000 .000 .000 .000 .000 .000 .045 .000 .000

227 Appendix 7. Hardy-Weinberg Equilibrium probabilities for all populations estimated by the Markov chain method for loci with more than five alleles. Populatioins are as in Table 6.

1.3a 3.1.2b 3.3.1b 3.3.2b 4.4.2b 4.4.10b Over All Loci Tiger Quoll MW .0251- 1+ 1+ .5427- .5693+ .4659- .5077- GI --- 1+ .2381- 1- .3333- 1- .8866- CO .3846- .3412- .5874+ .0775- .8138+ .2482- .3380- CH .7071- .8426- .4099- .0824+ .3278- 1+ .6121- BG 1- .5343- .5217- .8083- .3795- .6312+ .9240- BT 1- .0938- .7028+ 1+ .7523- .4118- .8019- BS 1- .0303- --- .2381- .6364+ 1+ .3759- SB .4667------1- 1+ 1+ .9923+ WY .0590- 1+ --- .1919- .0015- .3333- .0070- TT 1+ .4213- .0610- .4056- .7561- .4157+ .4915-

Eastern Quoll NE .0112- .0000- .1253- .1241- .0000- .3333- HS- ST .1674- --- .0043- .0667------.0029- GL .8288- .0170- .0000+ .0000+ .3846- .4097- HS+ VB .1847- 1- .0744- .0013- .5620+ .0762- .0052- TE .1945- 1+ .7473- .0158- .0261- .9119- .0743-

Northern Quoll KA .8344+ .0040- .6823+ .0008- .0712- .1714+ .0004- AR 1+ .2000- .3143= --- .3291+ --- .4576- AT --- .2496- .1887- .6536+ .1712- --- .2323-

Western Quoll PE .0016- .2279+ .0000- .0723- .9306+ .3101- HS- BF .0007- .0021- .0032+ .0010- .0002- .0073- .0000- HS highly significant; - heterozygote deficit + heterozygote excess = neither heterozygote excess or deficit a locus derived from eastern quoll DNA b loci derived from tiger and eastern quoll DNA

228