The role of odour in Australian mammalian

predator/prey interactions

Benjamin Gallard Russell B.Sc.

A thesis submitted in fulfilment

of the requirements for the degree of

Doctor of Philosophy

School of Biological, Earth and Environmental Sciences

University of New South Wales

Sydney NSW Australia

December 2005 Originality Statement

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

Signed …………………………………………………. Abstract i

Abstract

Odour plays an important role in many predator/prey interactions. In the northern hemisphere, many mammalian prey species have been shown to respond to predator odours. It is also widely assumed that mammalian predators utilise odours to locate their prey. This thesis explores the importance of odour in Australian mammalian predator/prey interactions.

Responses of native Australian species to the faecal odour of two predators; the native tiger quoll Dasyurus maculatus and the introduced red fox Vulpes vulpes, were evaluated through live-trapping and focussed behavioural studies of captive animals.

Tiger quoll responses to prey olfactory cues were investigated in a captive experiment.

Native rodents (bush rats Rattus fuscipes, swamp rats R. lutreolus and eastern chestnut mice Pseudomys gracilicaudatus) equally avoided traps scented with either quoll or fox faeces, and in captive experiments, bush rats and swamp rats reduced their average speed in response to both predator odours. Of the marsupial species, northern brown bandicoots Isoodon macrourus and common brushtail possums Trichosurus vulpecula were captured more frequently in quoll-scented traps than unscented traps or fox- scented traps, while captures of brown antechinus Antechinus stuarttii, long-nosed bandicoots Perameles nasuta and southern brown bandicoot I. obesulus were unaffected by the either predator odour. In captive experiments, brown antechinus, long-nosed and northern brown bandicoots decreased their foraging in response to both predator odours, and spent less time in areas scented with quoll faeces. Tiger quolls didn’t appear to detect odour sources from a distance of >65 cm, but they did follow Abstract ii

scent trails and spent more time in areas scented with the urine and faeces of potential prey.

Chemical analysis revealed no common components in fox and quoll odour which prey species could be responding to. Therefore, these native species have evolved to respond to fox odour since foxes were introduced to Australia 130 years ago. The stronger response of native rodents to fox odour may be a legacy of their co-evolution with canid predators prior to entering Australia. A better understanding of how odour is utilised in Australian predator/prey interactions may lead to a greater ability to protect

Australia’s unique mammalian fauna from introduced predators. Acknowledgements iii

Acknowledgements

There are a great many people I need to thank for helping me to be able to finish this thesis. But I’ll start at the beginning. Karen Firestone was the person who originally convinced me I should do a PhD, and for that I am extremely grateful. Chris Dickman suggested that Peter Banks would be the ideal person to supervise this thesis, and that he was likely to be appointed to the position at UNSW vacated by my honours supervisor Mike Augee. When that did occur, Barry Fox also recommended Peter as an ideal supervisor for what I wanted to do, and recommended myself as a potential PhD student to Peter. I would like to thank both Chris and Barry for helping to facilitate the beginning of my PhD journey.

To my supervisor Peter Banks, what can I say; thank you for putting up with me for all this time. You took on a boofy blokey naïve young kid as a PhD student all those years ago, when you were fresh off the plane from Finland, before you had even settled in to your new position. I hope I haven’t been too much of a burden. You’ve always been there with help and advice, and put up with my many questions, arguments, quirks and foibles. I couldn’t have done it without you and am tremendously appreciative of your support.

I’d also like to thank Ross McMutrie for his help and support over the years, in his role as postgraduate coordinator here at the School of BEES.

This thesis would not have been possible without people supplying me with plenty of carnivore urine and faeces. So I definitely need to send a big vote of thanks to Brad Walker and Chad Staples from Featherdale Wildlife Park, for helping me to collect tiger quoll urine and faeces, Steve Henry at the CSIRO sustainable ecosystems unit for collecting fox faeces for me, and Roy Bladen for collecting fox bladders filled with urine from animals he shot in western NSW.

This thesis received constructive criticisms from a number of people, which definitely led to an improvement in its quality. Obviously, the comments from my supervisor Acknowledgements iv

Peter Banks were invaluable. I would also like to thank Trent Penman for rapidly reading and reviewing every chapter, and getting them back to me within days (and sometimes hours) of me sending them to him. Similarly, I would like to thank Bruce Mitchell who quickly and thoroughly scrutinized the first four chapters, before being overwhelmed by the birth of his second child Hamish. The following people also provided insights into various chapters in the thesis; Barry Fox, Nelika Hughes, Jennifer Kelley, Adam Munn, Alan Russell, Diane Russell, and Jonathan Russell.

A number of people made my time conducting field work easier and more enjoyable and I would like to thank the following volunteers: Callum Juniper, Megan Lenardon, Amy Plunkett-Cole, Alan Russell, Tim Chapman, Jeremy Green, Rachel Miller, George Madani, Leanne Van Der Weyde, Michael Whitehead, Suzi Lewis, Raquel Melendez, Brian Hawkins, Dan Ramp, Matt Hayward, Ben Macdonald, Bruce Mitchell, Nelika Hughes, Jamie Russell, and Jonathan Russell. I would also like to thank Jan Nedved, Jeff Vaughn, Mark Russell and Diane Russell for their help with the captive experiments conducted at Cowan.

Thanks to all of the Friday night pub crew for helping me to relax and unwind at the Royal at the end of the week, while still providing entertaining insights, both biological and otherwise. Particularly I’d like to thank the regulars: the originals, Troy Gaston, Sharon Longford, Stephanie Moore and Matty Hayward; the next generation, Dustin Marshall, Jon Evans, and Jen Kelley; and those of us who remain, Keyne Munro, Stephen Bonser, Tom Mullaney, Jason Everett, Nelika Hughes, Candida Barclay and Frank Hemmings. And thanks to Macca, Wombat, Horatio, Dodge and Scooter for providing the geographers’ point of view during many nights at the Unibar over a few schooners.

I’d also like to thank the various members of the Banks’ Lab for putting up with me over the years. I’d particularly like to thank Fiona Powell, Nelika Hughes, Bruce Mitchell, Candida Barclay, Tania Rose (and Adam Munn), Jenna Bytheway, and Rochelle Basham for listening to my ramblings, giving me advice, coping with the mess created in my corner of the lab, and dealing with the odour from my staple diet of Irish Breakfast Tea, green cordial, vegemite toast and jaffles. Thanks guys. Acknowledgements v

Finally, my family has had to put up with a lot from me over the course of this PhD for which I need to thank them. They’ve dealt with the good times and bad, the anger and depression, the overriding exhaustion, they’ve allowed me to borrow and swap cars when I’ve needed to, and dealt with the backyard being covered in Elliot traps in various states of cleaning, and finding plastic bags filled with scats in the garage fridge. Despite this, my brothers; Jamie, Mark and Jonathan have all helped me to conduct my experiments, and have endlessly kept my spirits up with their banter and camaraderie that only brothers can share. My parents; Alan and Diane have also helped me to both conduct experiments and read all that I have written. Their love and support, encouragement and acceptance of what I have needed to do has never wavered, and without them I have no doubt that I would not have been able to finish this thesis. I am eternally grateful for everything they have done for and given to me, and love them very much.

And I also have to thank God, for giving us this amazing complex world we live in, which I spend every day being amazed by, where even urine and faeces are fascinating enough to base an entire PhD thesis around. Table of Contents vi

Table of Contents

Abstract ...... i

Acknowledgements...... iii

Table of Contents...... vi List of Tables...... x List of Figures...... xii

Chapter 1: Introduction...... 1 1.1 Species interactions ...... 2 1.2 Predation...... 2 1.3 Predator/prey arms race...... 4 1.4 Use of olfaction by mammal species...... 5 1.5 Prey responses to predator odours...... 5 1.5.1 Spatial responses...... 6 1.5.2 Foraging responses ...... 7 1.5.3 Locomotory responses...... 9 1.5.4 Life history responses...... 10 1.6 Continuity of prey responses...... 11 1.7 Specificity of prey responses...... 12 1.8 Use of olfaction by predators ...... 15 1.9 Odour mediated predator/prey interactions in Australia ...... 18 1.10 Aims and Structure of this thesis...... 22 1.11 References ...... 24

Chapter 2: Study Species ...... 33 2.1 Predator species...... 33 2.1.1 Australia’s mammalian predators...... 33 2.1.2 The Tiger Quoll ...... 36 2.1.3 The Red Fox...... 40 2.2 Prey species ...... 44 2.2.1 The Brown Antechinus...... 44 2.2.2 The Native Rodents...... 46 2.2.3 The Bandicoots ...... 52 2.2.4 The Common Brushtail Possum ...... 57 2.3 References ...... 59

Chapter 3: Responses of four Critical Weight Range (CWR) marsupials to the odours of native and introduced predators...... 68 3.1 Abstract...... 68 3.2 Introduction ...... 69 3.3 Methods ...... 72 3.4 Results ...... 74 Table of Contents vii

3.5 Discussion...... 76 3.6 Acknowledgements ...... 80 3.7 References ...... 81

Chapter 4: Do Australian small mammals avoid native and introduced predator odours? ...... 85 4.1 Abstract...... 85 4.2 Introduction ...... 86 4.3 Study area ...... 89 4.4 Methods ...... 90 4.4.1 Sampling designs...... 91 4.4.2 Data analysis...... 92 4.5 Results ...... 93 4.5.1 Bush Rats...... 94 4.5.2 Swamp Rats ...... 94 4.5.3 Eastern Chestnut Mice ...... 94 4.5.4 Brown Antechinus...... 95 4.5.5 Weights ...... 95 4.6 Discussion...... 100 4.7 Conclusion...... 106 4.8 Acknowledgements ...... 106 4.9 References ...... 108

Chapter 5: Behavioural responses of Australian small mammals to the faecal odours of native and introduced predators...... 112 5.1 Abstract...... 112 5.2 Introduction ...... 113 5.3 Methods ...... 117 5.3.1 Animals...... 117 5.3.2 Enclosures ...... 118 5.3.3 Trial procedure...... 119 5.3.4 Data analysis...... 121 5.4 Results ...... 122 5.4.1 Bush Rats...... 122 5.4.2 Swamp Rats ...... 123 5.4.3 Brown Antechinus...... 123 5.5 Discussion...... 129 5.5.1 Foraging responses ...... 129 5.5.2 Spatial responses...... 131 5.5.3 Locomotory responses...... 132 5.5.4 Scalar effects on locomotory responses ...... 133 5.5.5 Anti-predator behaviours revealed by trapping vs behavioural observations...... 134 5.5.6 Rodent responses vs antechinus responses...... 136 5.5.7 Conclusions ...... 136 5.6 Acknowledgements ...... 138 5.7 References ...... 139 Table of Contents viii

Chapter 6: Anti-predator strategies of Australian bandicoot species in response to olfactory cues from native and introduced predators...... 144 6.1 Abstract...... 144 6.2 Introduction ...... 145 6.3 Methods ...... 148 6.3.1 Animals...... 148 6.3.2 Enclosure...... 148 6.3.3 Trial procedure...... 150 6.3.4 Data analysis...... 151 6.4 Results ...... 152 6.4.1 GUDs...... 152 6.4.2 Time spent in the odour-treated pen...... 152 6.4.3 Number of times the odour-treated pen was entered...... 153 6.4.4 Average speed...... 153 6.5 Discussion...... 159 6.5.1 Foraging responses ...... 160 6.5.2 Movement responses...... 161 6.5.3 Conservation implications...... 162 6.6 Acknowledgements ...... 164 6.7 References ...... 165

Chapter 7: Comparisons between the volatile constituents of red fox Vulpes vulpes and tiger quoll Dasyurus maculatus urinary and faecal odours...... 169 7.1 Abstract...... 169 7.2 Introduction ...... 169 7.3 Methods ...... 173 7.4 Results ...... 174 7.5 Discussion...... 179 7.6 Acknowledgements ...... 181 7.7 References ...... 182

Chapter 8: The role of odour in the prey searching behaviour of tiger quolls Dasyurus maculatus ...... 186 8.1 Abstract...... 186 8.2 Introduction ...... 187 8.3 Methods ...... 189 8.3.1 Experimental enclosure...... 190 8.3.2 Trial procedure...... 190 8.3.3 Data analysis...... 193 8.4 Results ...... 194 8.5 Discussion...... 198 8.6 Acknowledgements ...... 203 8.7 References ...... 204

Chapter 9: Conclusions and Implications ...... 207 9.1 Summary of results...... 207 9.1.1 Native Rodents...... 207 Table of Contents ix

9.1.2 Brown Antechinus...... 207 9.1.3 Bandicoots ...... 208 9.1.4 Common Brushtail Possums...... 208 9.1.5 Tiger Quolls...... 208 9.2 Synthesis of prey responses to predator odours ...... 210 9.3 Scalar effects on odour signal exploitation ...... 214 9.4 Olfaction vs other senses used in predator avoidance...... 217 9.5 Implications of olfactory signal exploitation to conservation ...... 219 9.6 References ...... 225

Appendix 1: A consideration of odour: how do mammals utilise their olfactory capabilities?...... 233 A1.1 Olfactory perception...... 233 A1.2 Odour plume dynamics...... 234 A1.3 Location of odour sources by insects ...... 235 A1.4 Location of odour sources by mammals...... 236 A1.5 Odour discrimination...... 239 A1.6 Olfactory cues and predation...... 240 A1.7 Conclusion...... 243 A1.8 References ...... 244 List of Tables x

List of Tables

Table 1.1: Studies which have found a reduction in feeding in response to predator odours ...... 8 Table 1.2: Studies which have investigated whether giving up densities (GUDs) are affected by predator odours...... 9 Table 4.1: Comparison of the weights of northern brown bandicoots and common brushtail possums versus the weights of rodent species which have exhibited predator odour avoidance in trapping studies...... 88 Table 4.2: Results of unreplicated ANOVAs comparing the number of captures per site in unscented, fox-scented and quoll-scented traps of four species of small mammal, using the single trap and three trap sampling designs...... 95 Table 4.3: Results of Binary Logistic Regressions comparing the number of captures in unscented, fox-scented and quoll-scented traps between the three trap and single trap sampling designs, and (for bush rats and brown antechinus) between locations (Myall Lakes or Sydney) for four species of Australian small mammal...... 98 Table 4.4: Results of two-way ANOVAs comparing the weights of animals caught in unscented, fox-scented and quoll-scented traps for bush rats, swamp rats, eastern chestnut mice and brown antechinus...... 99 Table 5.1: Results of two-factor repeated measures ANOVAs to test whether bush rats respond to tiger quoll and red fox faecal odour by altering: (a) their GUDs in the odour treated enclosure; (b) the total amount of time they spent in the odour- treated enclosure; (c) the number of times they entered the odour treated enclosure; (d) their average speed of movement in the odour treated enclosure; (e) their average speed of movement in the untreated feeding enclosure...... 126 Table 5.2: Results of two-factor repeated measures ANOVAs to test whether swamp rats respond to tiger quoll and red fox faecal odour by altering: (a) their GUDs in the odour treated enclosure; (b) the total amount of time they spent in the odour- treated enclosure; (c) the number of times they entered the odour treated enclosure; (d) their average speed of movement in the odour treated enclosure; (e) their average speed of movement in the untreated feeding enclosure...... 127 Table 5.3: Results of two-factor repeated measures ANOVAs to test whether brown antechinus respond to tiger quoll and red fox faecal odour by altering: (a) their GUDs in the odour treated enclosure; (b) their GUDs in the untreated feeding enclosure (c) the total amount of time they spent in the odour-treated enclosure; (d) the number of times they entered the odour treated enclosure; (e) their average speed of movement in the odour treated enclosure...... 128 Table 5.4: Summary of the responses of three species of Australian small mammals to the faecal odour of two predators, the native tiger quoll and the introduced red fox...... 129 Table 6.1: Results of three-factor repeated measures ANOVAs to test whether northern brown bandicoots and long-nosed bandicoots altered: (a) their GUDs in the odour- treated pen; (b) their GUDs in the untreated feeding pen...... 154 Table 6.2: Results of three-factor repeated measures ANOVAs to test whether northern brown bandicoots and long-nosed bandicoots altered the total amount of time they spent in the odour-treated pen...... 156 Table 6.3: Results of three-factor repeated measures ANOVAs to test whether northern brown bandicoots and long-nosed bandicoots altered the total number of times they entered the odour-treated pen...... 157 List of Tables xi

Table 6.4: Results of three-factor repeated measures ANOVAs to test whether northern brown bandicoots and long-nosed bandicoots altered their average speed in the odour-treated pen...... 158 Table 6.5: Summary of the responses of northern brown bandicoots and long-nosed bandicoots to the faecal odour of two predators; the native tiger quoll and the introduced red fox...... 159 Table 7.1: Common volatile components of fox and quoll faeces and urine, and their peaks as numbered in figs 7.1-7.4 ...... 176 Table 9.1: Summary of the responses of eight Australian prey species to the faecal odour of two predators, the native tiger quoll and the introduced red fox. Where - = no change, p = decrease, and n = increase. N/A indicates this response was not investigated for these species...... 209 Table 9.2: Maximum distance over which odour sources could be detected by mammalian predators in conditions with a wind speed of less than 10 km/h...... 217 List of Figures xii

List of Figures

Figure 2.1: The tiger quoll Dasyurus maculatus and its distribution in Australia (after Strahan 1994)...... 37 Figure 2.2: The red fox Vulpes vulpes and its distribution in Australia (after Strahan 1994)...... 41 Figure 2.3: The brown antechinus Antechinus stuartii and its distribution in Australia (after Strahan 1994)...... 45 Figure 2.4: The bush rat Rattus fuscipes and its distribution in Australia (after Strahan 1994)...... 48 Figure 2.5: The swamp rat Rattus lutreolus and its distribution in Australia (after Strahan 1994)...... 49 Figure 2.6: The eastern chestnut mouse Pseudomys gracilicaudatus and its distribution in Australia (after Strahan 1994)...... 51 Figure 2.7: The northern brown bandicoot Isoodon macrourus and its distribution in Australia (after Strahan 1994)...... 53 Figure 2.8: The southern brown bandicoot Isoodon obesulus and its distribution in Australia (after Strahan 1994)...... 54 Figure 2.9: The long-nosed bandicoot Perameles nasuta and its distribution in Australia (after Strahan 1994)...... 55 Figure 2.10: The common brushtail possum Trichosurus vulpecula and its distribution in Australia (after Strahan 1994)...... 58 Figure 3.1: The number of long-nosed bandicoots, southern brown bandicoots, northern brown bandicoots and common brushtail possums caught in unscented traps, traps scented with fox faeces, and traps scented with quoll faeces. * represents a significant difference from random entry of traps at the P=0.05 level...... 76 Fig 4.1: The mean number (± standard error) of bush rats, swamp rats, eastern chestnut mice, and brown antechinus caught per site in unscented, fox-scented and quoll- scented traps using (a) the three trap design, and (b) the single trap design. Different letters above the bars for each species indicate a significant difference (P<0.05). c* indicates that although not significant there was a strong trend towards a difference (P=0.061)...... 97 Fig 4.2: The mean weight in grams (± standard error) of bush rats, swamp rats, eastern chestnut mice, and brown antechinus caught in unscented, fox-scented and quoll- scented traps...... 99 Figure 5.1: Layout of interconnected experimental enclosures...... 119 Figure 5.2: Mean ± SE change in giving up densities (GUDs) of bush rats, swamp rats and brown antechinus between day 2 (before the predator odour was introduced) and day 3 (when the odour was introduced) in the enclosure where the odour was presented...... 124 Figure 5.3: Mean ± SE change in the amount of time bush rats, swamp rats and brown antechinus spent in the enclosure where the odour was presented, between day 2 (before the odour was introduced) and day 3 (when the odour was introduced). 124 Figure 5.4: Mean ± SE change in the number of times bush rats, swamp rats and brown antechinus entered (and left) the enclosure where the odour was presented, between day 2 (before the odour was introduced) and day 3 (when the odour was introduced)...... 125 Figure 5.5: Mean ± SE change in the average speed of movement of bush rats, swamp rats and brown antechinus in the enclosure where odour was presented, between List of Figures xiii

day 2 (before the odour was introduced) and day 3 (when the odour was introduced)...... 125 Figure 6.1: Plan of experimental enclosure...... 150 Figure 6.2: Mean ± SE change in giving up densities (GUDs) of northern brown bandicoots and long-nosed bandicoots between day 2 (before the odour was introduced) to day 3 (after the odour was introduced) in: (a) the pen where the odour was presented; and (b) the untreated feeding pen...... 155 Figure 6.3: Mean ± SE change between day 2 (before the odour was introduced) and day 3 (after the odour was introduced) in the amount of time northern brown bandicoots and long-nosed bandicoots spent in the pen where the odour was presented...... 156 Figure 6.4: Mean ± SE change between day 2 (before the odour was introduced) and day 3 (after the odour was introduced) in the number of times northern brown bandicoots and long-nosed bandicoots entered the pen where the odour was presented...... 157 Figure 6.5: Mean ± SE change between day 2 (before the odour was introduced) and day 3 (after the odour was introduced) in the average speed of northern brown bandicoots and long-nosed bandicoots in the pen where the odour was presented...... 158 Figure 7.1: Chromatogram of the headspace volatile compounds from red fox urine. Major peaks are numbered 1-17...... 176 Figure 7.2: Chromatogram of the headspace volatiles compounds from tiger quoll urine. Major peaks are numbered 1-23...... 177 Figure 7.3: Chromatogram of the headspace volatile compounds from red fox faeces. Major peaks are numbered 1-17...... 177 Figure 7.4: Chromatogram of the headspace volatile compounds from tiger quoll faeces. Major peaks are number 1-24...... 178 Figure 7.5: Two dimensional MDS ordination of odour components of red fox and tiger quoll urine and faeces, based on a square root transformation and a Bray- Curtis similarity matrix. All 24 samples are shown, six for each group, however for the same group most are clustered at exactly the same point...... 178 Figure 8.1: Experimental apparatus with tiger quoll (a) climbing central ramp, and (b) approaching platform...... 191 Figure 8.2: The mean difference ± SE between the time (in seconds) taken by the tiger quolls to reach the odour platform from the top of the ramp, and the time taken to reach the other platform from the top of the ramp when there were (a) no trails, (b) odour trails added, (c) odour trails added. Negative values indicate less time taken to reach the odour platform...... 196 Figure 8.3: The mean difference ± SE between the time (in seconds) spent by the tiger quolls on the odour platform and the time spent at the other platform when there were (a) no trails, (b) odour trails added, (c) odour trails added. Positive values indicate more time spent at the odour platform...... 197 Chapter 1: Introduction 1

Chapter 1:

Introduction

The role of odour in mammalian predator/prey interactions has become increasingly studied over the past 30 years in Europe and North America. This is an area of research which has received only limited attention in Australia, and has been focussed on how native prey species respond to the odours of introduced predators. The evolutionary history of Australia’s mammalian fauna allows for a unique opportunity to examine how prey species respond to the odours of predators with which they are evolutionarily unfamiliar. But in order to properly interpret these responses it is necessary to understand the role of odour in the interactions between native predators and prey.

This thesis addresses this issue and explores the importance of odour in the interactions between mammalian predators and prey in Australia. The main questions addressed are how native prey species respond to olfactory cues from both native and introduced predators, and also whether native predators exploit olfactory cues from native prey species.

The following introductory chapter briefly reviews what is known about the role of odour in mammalian predator/prey interactions; beginning with the definition of predation within the context of species interactions; and then outlining the results and implications of the research which has already been undertaken in the major aspects of this field of study, including the few studies from Australia. This is followed by why

Australia’s mammalian fauna is so different to other parts of the world, and the Chapter 1: Introduction 2

implications of understanding how Australian predators and prey respond to potential olfactory cues. The chapter concludes with the specific aims and structure of the thesis.

1.1 Species interactions

When grown under controlled laboratory conditions, the growth of simple organisms such as yeast or paramecium can be described by the logistic equation, where population growth is controlled solely by a limiting resource such as space or food

(Krebs 1994). However, most populations in the world do not exist as discrete individual entities, but rather as part of a much larger complex community structure.

The nature of a given population in such a community is affected by its interactions with the other species in that community (Stiling 2002). There are five basic types of species interaction based on whether the interaction has a positive, negative or neutral effect on each of the two species (Begon et al. 1996b). When both species are positively affected it is known as mutualism; when one benefits whilst the other remains unaffected it is known as commensalism; competition is when both species are affected negatively; amensalism is when one species is negatively affected whilst the other is unaffected; but the interaction which has a positive effect on one species and a negative effect on the other is known as predation (Begon et al. 1996b).

1.2 Predation

Predation is the consumption of one organism (the prey) by another (the predator), where living prey are attacked by the predator prior to consumption (Begon et al.

1996a). This broad general definition includes parasites and parasitoids, which slowly consume their host while the host is still alive, as well as herbivores, which consume plant material. But it is carnivores, those animals which hunt, kill and consume other Chapter 1: Introduction 3

animals which are most commonly thought of when the word predator is mentioned

(Krebs 1994; Stiling 2002). As prey species are the ones to which the interaction of predation has a negative effect, different prey species have developed a wide range of anti-predation strategies. These include (Stiling 1992; Stiling 2002):

x Aposematic or warning colouration: colouration which warns potential

predators that the displayer is either unpalatable or noxious.

x Crypsis and catalepsis: blending into the background and remaining still to

prevent detection by predators.

x Mimicry: mimicking of another animal which may be dangerous or unpalatable

to predators.

x Intimidation: display to deceive predators into thinking the prey will be more

difficult to subdue.

x Polymorphism: the occurrence within a population of two or more discrete

forms of a species in proportions greater than can be maintained by recurrent

mutation alone. This is thought to confound predator search image.

x Phenological separation: changes in a species activity period in different

locations due to the activity period of the predators in the area.

x Chemical defences: when attacked some prey species will employ noxious

chemicals against the predator.

x Masting: the synchronous production of more progeny than can be consumed by

predators, ensuring that at least some survive

Similarly, predators have evolved a range of methods for hunting prey. In order to encounter prey some predators systematically search, while others lie in wait. Once prey are located some predators rely on ambush tactics, some approach slowly, stalking Chapter 1: Introduction 4

their prey, whilst others will openly approach prey and run them down. The actual hunting method may be a combination of those described above, the same predator may employ different tactics against different prey species, or even different tactics against the same species depending on current conditions (Curio 1976).

1.3 Predator/prey arms race

The strategies of predators for hunting prey, and of prey for escaping predation are not static, but rather are constantly evolving in a dynamic “arms race” (Dawkins and Krebs

1979). As predators evolve adaptations to become ever more efficient hunters, prey species evolve ever better techniques to avoid being preyed upon. However, deployment of these techniques may not be without some cost. Whilst individual prey avoid the ultimate fitness cost of death, these behavioural changes can have “sublethal” effects on the population (Lima 1998). Reproduction and long-term survival may be negatively affected as opportunities to feed, socialise and copulate are compromised by the trade-offs with the avoidance of predation (Lima and Dill 1990; Ronkainen and

Ylönen 1994). Because of this, many anti-predator strategies are only implemented under a perceived level of predation risk, as assessed by remote sensory cues (Lima and

Dill 1990). Similarly, predators change their hunting techniques based on sensory cues associated with potential prey species. Smith (1974) described the “search strategy” as the general search of a predator for cues indicating a specific distribution of prey, and described the “search tactic” as an adaptive change in behaviour once the predator arrived in an area where such cues indicated high prey density.

These cues may be indirect, such as habitat changes which affect the vulnerability of prey to predators, but indirect cues overestimate risk when predators are absent (Sih Chapter 1: Introduction 5

1992; Thorsen et al. 1998), and often changes in behaviour will only result from direct evidence, be it visual, aural or olfactory (Kats and Dill 1998). Visual and aural cues are generally instantaneous, providing information about the actual presence of an animal at a specific point in time (Jones et al. 2004). As such, responses to these cues, although immediate, tend to only be short term responses. Olfactory cues on the other hand are indicative of the recent presence of an animal. In essence they represent an increased likelihood of encountering an animal, rather than the actual presence of that animal.

Olfactory cues often last a lot longer than visual and aural cues, as do the responses to olfactory cues (Kats and Dill 1998).

1.4 Use of olfaction by mammal species

Olfaction plays a prominent role in the lives of most mammal species (Stoddart 1976b).

Odour is used in various social interactions, conveying information about sex, reproductive and emotional state, and group and individual identity (Brahmachary

1986). Many mammal species also mark out their territories with a combination of urine, faeces, and odorous secretions of specialized skin glands (Macdonald and Brown

1985). Such scent marks often produce long-lasting signals which persist in the environment (Regnier and Goodwin 1976). But being open signals rather than strongly directional, the information they contain is potentially not only communicated to conspecifics, but may also be exploited by unintended “eavesdroppers” (McGregor

1993; Roberts et al. 2001).

1.5 Prey responses to predator odours

Mammalian prey species readily eavesdrop upon such signals left by their predators, and exploit this information to decrease their risk of predation. Differing responses to Chapter 1: Introduction 6

predator odours have been recorded in a wide range of mammalian prey species, with over 100 studies published on the subject (Kats and Dill 1998). These responses may be divided up into four broad categories: spatial responses, foraging responses, locomotory responses, and life history responses (Ylönen 2001).

1.5.1 Spatial responses

Probably the most obvious response to the predation risk indicated by predator odours is to avoid the area where the olfactory cue is encountered (Jedrzejewski and

Jedrzejewska 1990). Carnivores regularly revisit the scent marks they deposit, sometimes on a daily basis, and often remark the same area on subsequent visits

(Macdonald 1980). These scent marks may also be investigated by conspecifics

(Gorman 1980; Macdonald 1980; Gorman and Trowbridge 1989), so by avoiding these areas, prey individuals reduce their likelihood of encountering these predators.

Field support for this basic hypothesis has come from several trapping studies using the

“scat at trap” technique (sensu Powell and Banks 2004). Captures of field voles

Microtus agrestis were significantly decreased when traps were scented with the anal gland secretion of weasels Mustela nivalis (Stoddart 1976a) and stoats M. erminea

(Gorman 1984), and captures of woodmice Apodemus sylvaticus and bank voles

Clethrionomys glareolus were lower in traps scented with the faecal odour of the red fox Vulpes vulpes than in traps scented with rodent faecal odour or unscented traps

(Dickman and Doncaster 1984). However, such responses are by no means universal. In the same studies mentioned above, captures of woodmice were unaffected by either mustelid odour (Stoddart 1976a; Gorman 1984), and field voles did not avoid traps scented with red fox faecal odour (Dickman and Doncaster 1984). These differences in avoidance are unlikely to be simply due to differing degrees of predation pressure, as Chapter 1: Introduction 7

field voles are preyed on more heavily than woodmice by all three predator species

(Lever 1959; Day 1968; Moors 1975; Richards 1977; King 1980; Doncaster et al. 1990;

McDonald et al. 2000), in some areas despite the fact that wood mice are far more common (King 1980; Dickman and Doncaster 1987; Doncaster et al. 1990). King

(1990) suggested that rather than avoid traps scented with mustelid odours, woodmice might be responding to these odours with other behavioural changes such as a reduction in conspicuous movements. It is possible that field voles similarly respond to fox odour in a manner which does not reduce their capture rate.

1.5.2 Foraging responses

As well as spatial avoidance, the presence of predator odours can also lead to a reduction in other activities in “risky” areas, such as foraging. For example, cougar

Felis concolor, coyote Canis latrans and wolf C. lupus faeces, as well as coyote, wolf, red fox, wolverine Gulo gulo, lynx F. lynx and bobcat F. rufus urines significantly reduced browsing on conifer seedlings by black-tailed deer Odocoileus hemionus columbianus (Sullivan et al. 1985b) and snowshoe hare Lepus americanus feeding was decreased in the presence of lynx and bobcat faeces, weasel anal gland secretion, and lynx, bobcat, wolf, coyote, fox, and wolverine urines (Sullivan et al. 1985a). As some feeding, although greatly reduced, did occur in areas scented with these predator odours, these results do not necessarily represent complete avoidance of these areas, but rather a reduction in the amount of time spent in the vicinity of the odour source. The risk of predation is reduced, although not as greatly as complete avoidance, but this represents a trade-off between predation risk and acquisition of a resource, in this case food (Lima 1998). Other examples of reduced feeding in response to predator odours are shown in Table 1.1. Chapter 1: Introduction 8

Table 1.1: Studies which have found a reduction in feeding in response to predator odours Predator Odour Prey species Study

European hedgehogs (Ward et al. Eurasian badger Meles meles faeces Erinaceus europaeus 1997) Wolf and domestic dog Canis lupus (Arnould et al. Domestic sheep Ovis aries familiaris faeces 1998) Domestic cat Felis catus, red fox, (Carlsen et al. Field voles and American mink faeces 1999) (Bolbroe et al. Weasel faeces Field voles 2000) Red fox and racoon Procyon lotor Gray squirrels Sciurus (Rosell 2001) urine carolinensis

The foraging costs of predation risk have been frequently investigated via the use of artificial feeding patches to measure the quitting harvest rate or giving up density

(GUD) under varying levels of predation risk (Brown 1988). However, studies using

GUDs which have manipulated predator odours as a cue to predation risk have rarely found any foraging costs (Table 1.2). For example, while Dickman (1992) found that house mice Mus domesticus avoided traps scented with red fox faecal odour, Powell and Banks (2004) found that this response had no effect on their foraging behaviour as measured by GUDs. Of the five studies in Table 1.2, only two found any evidence of a trade-off between foraging and the predation risk represented by the predator odour.

Jones and Dayan (2000) found that Acomys cahirinus increased their GUDs in response to Blandford’s fox V. cahirinus faeces in one site, but not the other, in open microhabitats, but not in closed microhabitats, but they stated that the differences were subtle and only just reached significance at p=0.045. Herman and Valone (2000) found that in winter, but not in summer, kangaroo rats Dipdomys merriami foraged more in unscented food patches than food patches scented with gray fox Urocyon cinereoargenteus urine, although their results were only marginally significant at Chapter 1: Introduction 9

p=0.088. These results are somewhat equivocal, and no response was recorded in any of the other studies.

Table 1.2: Studies which have investigated whether giving up densities (GUDs) are affected by predator odours. Predator Odour Prey species Study

Fox squirrels Sciurus niger and thirteen- lined ground squirrels Spermophilus (Thorsen et al. Red fox urine tridecemlineatus 1998)

Spiny mice Acomys cahirinus and A. Blandford’s fox V. (Jones and russatus cahirinus faeces Dayan 2000) Gray fox Urocyon (Herman and cinereoargenteus urine Kangaroo rats Valone 2000) Bobcat, red fox, and (Orrock et al. coyote urine Oldfield mice Peromyscus polionotus 2004) (Powell and Red fox faeces House mice Banks 2004)

1.5.3 Locomotory responses

Although many prey species avoid areas where predator odours indicate increased predation risk, those animals which are found in these areas often reduce their locomotory activity or mobility in response to this increased predation risk. Freezing or creeping behaviours are presumed to reduce visual and aural cues of prey location, and are therefore most effective against predators which rely on these cues to locate prey

(Kats and Dill 1998). For example, in lab trials using small chambers (60 cm [L] x 26 cm [W] x 36 cm [H]) although brown rats Rattus norvegicus spent less time in areas tainted with the odour of collars from domestic cats Felis catus, when they were in the immediate vicinity of the cat collar they significantly reduced their locomotory activity

(Dielenberg et al. 2001; Dielenberg and McGregor 2001; McGregor et al. 2002). Chapter 1: Introduction 10

Similarly in larger enclosures (1.70 m [L] x 1.15 m [W] x 1.10 m [H]) fewer bank voles were found in areas tainted with predator odour, but those which were found in these areas reduced their locomotory activity in the presence of weasel and red fox odour

(Jedrzejewski et al. 1993). On a broader scale Norrdahl and Korpimäki (1998) found that mobility of field voles and sibling voles M. rossiaemeridionalis was significantly associated with predation risk; a reduction in predation risk resulted in increased mobility, and voles which were preyed upon by weasels and stoats were more mobile than voles which survived (Norrdahl and Korpimäki 1998).

1.5.4 Life history responses

One of the more subtle ways prey species can respond to predator odours is by altering aspects of their life history. In lab trials, breeding in the field vole and bank vole was suppressed by the urinary and faecal odour of mustelids, probably due to females avoiding copulation under high predation risk (Ronkainen and Ylönen 1994; Ylönen and Ronkainen 1994; Koskela and Ylönen 1995; Mappes and Ylönen 1997). These results led to the development of the breeding suppression hypothesis (BSH), which states that because reproduction under predation risk causes significant survival costs for female voles and/or their offspring, voles should decrease these costs by delaying their breeding to the future when predation risk is lower (Mappes et al. 1998). But the validity of the BSH was called into question when these results failed to be replicated in large scale enclosure experiments. American mink faeces and urine did not affect the breeding of gray-tailed voles Microtus canicaudus in 0.2 ha enclosures (Wolff and

Davis-Born 1997; Jonsson et al. 2000), and weasel bedding soiled with faeces and urine did not alter the breeding of bank voles in 0.25 ha enclosures (Mappes et al. 1998). It was concluded that the earlier results from the laboratory studies were a methodological Chapter 1: Introduction 11

artefact, and that mustelid odours did not affect reproduction in the field (Mappes et al.

1998; Jonsson et al. 2000). However, more recently, in a true field experiment Fuelling and Halle (2004) verified the BSH for the first time under natural conditions. The application of weasel bedding soiled with faeces and urine resulted in an increase in the number of reproductively non-active adult female grey-sided voles Clethrionomys rufocanus in a wild population from the subalpine tundra of northern Norway (Fuelling and Halle 2004).

1.6 Continuity of prey responses

The issue of a continuity of response between lab, enclosure and field experiments is not confined to the BSH. A number of studies have found that although prey species may respond to predator odours in one experiment, this does not guarantee that when the variable being measured and/or the scale of the experiment are changed, a response to the odour of the same predator will still be observed. For example, Parsons and

Bondrup-Nielsen (1996) found that meadow voles M. pennsylvanicus responded to stoat odour under lab conditions, but when trapped in a 40 m x 30 m outdoor enclosure, they were captured as often in stoat scented traps as guinea pig scented and unscented traps. Wolff and Davis-Born (1997) trapped fewer gray-tailed voles M. canicaudus in traps scented with American mink faeces than traps scented with European rabbit

Oryctolagus cuniculus faeces and unscented traps in 45 m x 45 m enclosures. But in enclosures of the same size, application of American mink faeces and urine to the unmowed unclipped areas of these enclosures did not result in them vacating this tall grass habitat and moving into the mowed open section of the enclosure. In the same enclosures American mink urine and faeces did not affect the maximum distance moved by gray-tailed voles, nor did weasel bedding affect mean and maximum distance Chapter 1: Introduction 12

moved or home range size of bank voles in 50 m x 50 m enclosures in Finland (Jonsson et al 2000). Both studies concluded that earlier smaller-scale results where prey species did respond to predator odours were prone to artefacts, and therefore not indicative or representative of a field situation (Wolff and Davis-Born 1997; Jonsson et al. 2000).

However, fine scale responses are difficult to measure under field conditions and may miss the subtleties and immediacy of prey responses to predator odours. For example,

Borowski (1998b) found that in the field, weasel odour had no affect on centres of activity or probability of encountering free-living root voles M. oeconomus, but their home ranges did decrease in size, and laboratory experiments revealed that although distances kept from predator odours were no greater than distances from control odours, root voles significantly decreased their locomotory activity (Borowski 1998b). Whilst root voles were not changing their spacing behaviour to avoid encountering weasels, they were altering their behaviour to reduce their likelihood of detection. This result suggests that an integrated approach using both laboratory and field experiments may be needed to allow for a more complete understanding of how prey respond to predator odours.

1.7 Specificity of prey responses

Several experiments on animals living on remote off-shore islands question whether odour avoidance responses by prey species are inherent or whether they can be lost and regained depending upon exposure to predation. Orkney voles M. arvalis avoid the odour of stoats (Gorman 1984) and red foxes (Calder and Gorman 1991) despite having no contact with either predator species for over 3000 years, suggesting such responses may be inherent. But on off-shore Western Australian islands where no mammalian Chapter 1: Introduction 13

predators are present, house mice show no avoidance of red fox and domestic cat odour after only a few hundred years of isolation (Dickman 1992). On other West Australian islands, where only cats are present, house mice avoid both fox and cat odour and

Dickman (1992) suggested that this avoidance might be elicited by components common to the odours of these two species. Although stoats and red foxes are absent from the Orkneys, two other mammalian predators are found on the islands; otters

Lutra lutra (Calder and Gorman 1991) and domestic cats, with the latter known to prey heavily upon Orkney voles (Corbet and Wallis 1977). It is possible that Orkney voles are likewise responding to odour components common to these carnivore species

(Calder and Gorman 1991; Borowski 1998a).

Jedrzejewski et al. (1993) found that bank voles respond to the odours of six species of mammalian predator; weasels, stoats, polecats Mustela putorius, stone martens Martes foina, red foxes, and racoon dogs Nyctereutes procyonoides. For all six species, although a significant number of bank voles avoided the pen tainted by predator odour, those which remained displayed a combination of anti-predator behaviours, which while different for each species, were consistent for the hunting technique of that predator (Jedrzejewski et al. 1993). This suggests they respond to something unique in the odour of each predator, but does not preclude the possibility that there may be common components to these odours which identify them as carnivores, and trigger a generalised response (Dickman and Doncaster 1984).

It has been suggested that the repellency of predator odours to prey species might be based on the recognition of sulfurous compounds derived from the digestion of meat

(Nolte et al. 1994). The aversion by guinea pigs Cavia porcellus, house mice, deermice Chapter 1: Introduction 14

Peromyscus maniculatus and mountain beaver Aplodontia rufa to feeding in the presence of coyote Canis latrans urine was lessened when sulfurous compounds were chemically removed and when coyotes were fed on fruit instead of meat (Nolte et al.

1994). The volatiles from mustelid anal scent gland secretions are dominated by a series of such sulfur containing compounds (Crump 1980; Sokolov et al. 1980; Brinck et al. 1983) and fox urine has similarly been found to contain a number of sulfurous components (Jorgenson et al. 1978; Wilson et al. 1978; Bailey et al. 1980).

Sullivan (1986) tested whether any of the nine compounds identified from the volatile component of red fox urine (Jorgenson et al. 1978; Wilson et al. 1978; Bailey et al.

1980) reduced feeding in the snowshoe hare, and found that one chemical, 3-methyl-3- butenyl methyl sulfide, was as effective as whole fox urine. However, the two other structurally similar sulfurous compounds, 2-phenylethyl methyl sulfide and 3- methylbutyl methyl sulfide, had no effect (Sullivan and Crump 1986). Similarly,

Sullivan and Crump (1984) tested the feeding response of snowshoe hares to four sulfurous compounds isolated from mustelid anal scent glands (Brinck et al. 1983) and three structurally similar synthetic compounds, and found that only two of the compounds, 3-propyl-1,2-dithiolane and 2,2-dimethylthietane, significantly affected their feeding behaviour.

This indicates that anti-predator strategies are not initiated in response to sulfurous metabolites in general, but rather are triggered by specific sulfurous compounds.

However, these specific sulfurous compounds may be acting as common component triggers. Robinson (1990) found that feeding by the European rabbit was reduced in the presence of 2,2-dimethylthietane. Although this is the major component in American Chapter 1: Introduction 15

mink anal sac secretions, it is also found in anal sac secretions of stoats and ferrets

Mustela putorius furo (Brinck et al. 1983). Woodmice and bank voles also avoided traps scented with this chemical (Robinson 1990). A generalised response to this common mustelid trigger may explain why British water voles Arvicola terrestris avoid the odour of introduced American mink (Barreto and Macdonald 1999).

Similarly, 3-methyl-3-butenyl methyl sulfide, the component responsible for the aversiveness of red fox urine (Sullivan and Crump 1986), has also been identified in the urine of wolves (Raymer et al. 1984; Raymer et al. 1986), coyotes (Schultz et al. 1988), and domestic dogs (Schultz et al. 1985), as well as the anal sac secretions of American mink (Sokolov et al. 1980). It is possible this compound is present in the odour of a number of other carnivore species as well. A generalised response to such an odour component common to a large number of carnivore species would also explain why field voles avoid traps scented with tiger Panthera tigris urine (Stoddart 1982a;

Stoddart 1982b), red deer Cervus elaphus reduce their feeding in the presence of Lion

Panthera leo faeces (Abbott et al. 1990), and black-tailed deer eat less in the presence of lions, tigers and snow leopards Panthera unica with which they are allopatric, as well as their sympatric predators the coyote and the cougar (Muller-Scwarze 1972).

1.8 Use of olfaction by predators

Just as prey may exploit the social odours of their predators in order to avoid risky situations, predators may exploit social odours of their prey in order to increase foraging efficiency. But while it is widely assumed that mammalian predators hunt using olfaction, there are surprisingly few experimental studies to support this assumption. Cushing (1985) found that oestrous female prairie deermice Peromyscus Chapter 1: Introduction 16

maniculatus bairdi were more vulnerable to predation by weasels than dioestrous prairie deermice, and weasels showed a preference for the odour of the urine from oestrous over dioestrous prairie deermice (Cushing 1984). Since their publication, it is these papers which have been cited as proof that mammalian predators use olfaction in hunting prey. They were also a contributing factor in the formulation of the BSH

(Ylönen et al. 2003). However, while Ylönen et al. (2003) found that weasels were attracted to the urinary odours of field voles and bank voles, they did not detect any preference for female voles versus male voles, or any effect of reproductive state.

More generally, the exploitation of prey odours by mammalian predators is poorly understood. Lima (2002) noted that over the last 20 years the majority of research in predator/prey interactions has been focussed on prey behaviours, and study into predator behaviour has been limited. This bias is particularly evident in research into the role of odour in mammalian predator/prey interactions. Apart from the papers mentioned above, the only quantitative investigations concerning the use of olfaction by mammalian predators have been the consideration of the relative importance of vision, audition and olfaction in the near-location of prey. Vision was found to be the most important sense in coyotes (Wells 1978; Wells and Lehner 1978), but while Wells and Lehner (1978) considered audition more important than olfaction in a small windless enclosure, Wells (1978) considered olfaction more important than audition in a larger enclosure open to air movement. On the other hand, Österholm (1964) found audition to be the most important sense for the location of prey in the red fox, but recognised that their relative importance was dependent on the conditions, and that all three senses were used concurrently in a single integrated effort. All other information Chapter 1: Introduction 17

available on the use of olfaction in the location of prey by predators has been gathered from anecdotal reports.

However, there is indirect evidence that prey may be vulnerable to the exploitation of their social odours by predators. Radio-tracking of field voles and sibling voles revealed a u-shaped relationship between predation and mobility of field and sibling voles, with the most vulnerable individuals being those with the highest and lowest mobility (Banks et al. 2000). Their empirical model predicted that highly mobile individuals would have a higher likelihood of random predator encounter, while the least mobile individuals would be especially vulnerable due to an increased accumulation of urine and faeces which attract weasel predators (Ylönen et al 2003).

Further to this, the high predation rate upon reintroduced sibling voles was linked to their low mobility in the first three days after release, resulting in an accumulation of urine and faeces attractive to predators (Banks et al. 2002). The addition of extra urinary and faecal odours of field voles and sibling voles to an area has been shown to result in an increase in predation on these species by mammalian predators (Koivula and Korpimäki 2001). Some prey species may be sensitive to the increased predation risk presented by such an accumulation of scent marks, as the presence of ferret urine has been shown to reduce scent marking in dominant male house mice in the lab

(Roberts et al. 2001). However, in lab and enclosure experiments American mink urine did not appear to affect the scent marking behaviour of prairie voles M. ochrogaster and woodland voles M. pinetorum (Wolff 2004). Chapter 1: Introduction 18

1.9 Odour mediated predator/prey interactions in Australia

The contemporary suite of mammalian fauna in Australia presents a unique opportunity to examine the importance of odour in predator/prey interactions involving evolutionary unfamiliar predators and prey, which will have important conservation implications. Australia has a distinct mammalian fauna that evolved in relative isolation for around 40 million years (Strahan 1998). Unlike the rest of the world, where related species tend to fill the same niches in different locations, in Australia many of these niches are filled by unrelated marsupial species. For example, the grazers are members of the marsupial order Diprotodontia, rather than belonging to the Perrisodactyla,

Artiodactyla, or Lagomorpha; and the majority of predators are members of the marsupial order Dasyuromorphia rather than the placental order Carnivora (Macdonald

2001).

However, humans have introduced three placental carnivore species to Australia that have since become established across mainland Australia. The dingo Canis lupus dingo was introduced to Australia by Asian seafarers approximately 4000 years ago (Corbett

1985; Corbett 1995). Domestic cats arrived in Australia with the first fleet in 1788 and established feral populations shortly thereafter (Dickman 1996b; Abbott 2002). And after several failed attempts the red fox was “successfully” introduced in the 1870s

(Rolls 1969). All three placental predators now prey upon native mammals and pose significant conservation concerns (Dickman 1996a). It has been speculated that the introduction of dingoes led to the extinction of the Thylacine Thylacinus cynocephalus and the Tasmanian devil Sarcophilus harisii on the Australian mainland (Corbett 1995).

While dingoes no doubt had an impact on native prey species as well (Dickman 1996a), no extinctions have been directly attributed to their arrival (Morton 1990; Short et al. Chapter 1: Introduction 19

2002). However, Corbett (1995) suggests that European alterations to the landscape increased the vulnerability of native prey species to the dingo, contributing to the extinction of up to 10 medium-sized mammal species in central Australia since the

1930s. Cats have been implicated in the extinction of up to eight species of native

Australian rodent prior to 1900, and regional declines of a number of other species, although the evidence, while compelling, is largely circumstantial due to the timeframe involved (Dickman et al. 1993; Dickman 1996a; Dickman 1996b; Smith and Quin

1996). The more recent introduction of foxes, on the other hand, has been closely linked with the decline and local extinction of a suite of marsupial species including: the greater bilby Macrotis lagotis; western barred bandicoot Perameles bougainville; numbat Myrmecobius fasciatus; eastern bettong Bettongia gaimardi; brush-tailed bettong B. penicillata; burrowing bettong B. lesueur; rufous bettong Aepyprymnus rufescens; quokka Setonix brachyurus; and tammar wallaby Macropus tammar (Watts

1969; Christensen 1980; Southgate 1990; Short 1998; Richards and Short 2003). They have also been implicated with declines in conilurine rodents (Smith and Quin 1996).

Predation by both feral cats and red foxes is listed as a key threatening process under the Australian Commonwealth Environment Protection and Biodiversity Conservation

Act (1999). These species evolved in areas where odour has been shown to play an important role in mammalian predator/prey interactions, yet a question which has important implications is to what degree is odour utilised by Australian native mammal species in predator/prey interactions.

There have been several experiments attempting to address this question. However, the results have been equivocal. Preliminary research suggests some native Australian mammals may be naïve to the odours of novel placental predators whereas others may Chapter 1: Introduction 20

show avoidance. Trap success of two common Australian small mammal species, the agile antechinus Antechinus agilis and the bush rat R. fuscipes were not affected by faeces of the introduced red fox (Banks 1998), nor was trap success of bush rats affected by domestic dog faecal odour (Banks et al. 2003). In pen trials, tammar wallabies and red-necked pademelons Thylogale thetis did not alter their feeding behaviour in response to red fox, brown bear Ursus arctos and dingo faeces and domestic dog urine (Blumstein et al. 2002). It is difficult to determine whether these results conflict with the common constituents hypothesis, as it is not known how these prey species respond to the odours of native marsupial predators. In contrast, application of domestic dog urine to seedlings reduced browsing by swamp wallabies

Wallabia bicolor (Montague et al. 1990), and application of synthetic predator odours reduced browsing by common brushtail possums Trichosurus vulpecula (Woolhouse and Morgan 1995; Morgan and Woolhouse 1997), but this may have been due to reduced palatability rather than a perceived increase in the risk of predation (Jones and

Dayan 2000). Nonetheless, feeding in wild common brushtail possums was reduced by the presence of red fox urine and faeces (Gresser 1996). The only study to consider the odour of a native Australian predator found that western quoll Dasyurus geofroii faecal odour decreased the trap success of introduced house mice in Western Australia, but to a lesser extent than the faecal odour of domestic cats and red foxes (Dickman 1992).

But no studies have examined how Australian native mammals may utilise the odours of their native predators or how native mammalian predators may utilise the odours of their prey.

Australia’s marsupials evolved in isolation from the rest of the world for approximately

40 million years (Strahan 1998), and their utilisation of odour in predator/prey Chapter 1: Introduction 21

interactions may have developed quite differently compared to placentals. It is possible that predation pressure was not great enough for responses to mammalian predator odours to either develop or be sustained, as it has been regularly reported that during the Quaternary Australia has had comparably fewer large mammalian carnivores than in other parts of the world (Hecht 1975; Rich and Hall 1984; Flannery 1991; Flannery

1997). However, this view has recently been called into question (Wroe et al. 1999;

Wroe 2002; Wroe 2003; Wroe et al. 2003) (see Chapter 2). If responses to predator odours have evolved, it is possible that prey species are responding to the same odour components as their counterparts in other parts of the world, especially if these compounds are commonly derived from a meat-based diet (Nolte et al. 1994).

Conversely the large degree of evolutionary separation between the marsupial order

Dasyuromorphia and the placental order Carnivora may have led to the development of completely different odour components, or if there are common components these may not be the components which Australian marsupials have evolved to respond to.

The ancestors of Australia’s native rodents on the other hand evolved with placental carnivores, and entered Australia in two distinct waves: 5-7 million years ago, and less than a million years ago (Watts and Aslin 1981; Godthelp 1989) (see chapter 2).

Presumably, if they evolved responses to the odours of these ancestral placental carnivores, and if the odour components rodents respond to are simply metabolites derived from a carnivorous diet (Nolte et al. 1994), they would respond to these chemicals as readily in the odours of marsupial carnivores as in the odours of placental carnivores. If not, these rodents may or may not have evolved to respond to the odours of their new marsupial predators, and may or may not have retained their responses to placental carnivore odours. Chapter 1: Introduction 22

By investigating the similarities and differences in the responses of Australian native rodent and marsupial species to the odours of a native marsupial predator, and an introduced placental predator, and comparing the chemical profiles of these odours, we can determine which of these scenarios are likely to be correct. Each scenario has different conservation implications in terms of the vulnerability of these prey species to introduced placental predators, as does the degree to which olfaction is used by marsupial predators to locate prey.

1.10 Aims and Structure of this thesis

The aims of this study are to

x Determine if and how Australian marsupial prey species respond to the faecal

odour of the native marsupial predator the tiger quoll D. maculatus and the

introduced placental predator the red fox.

x Determine if and how Australian native rodent species respond to the faecal

odour of the tiger quoll and the introduced placental predator the red fox.

x Determine whether there are any common chemical components in the odour of

the tiger quoll and the red fox.

x Determine whether tiger quolls utilise prey species odours as cues for hunting.

This thesis consists largely of a series of stand-alone chapters, each with their own abstract, introduction, methods, results, discussion, acknowledgements, and references, which have been or will be modified and submitted to a scientific journal for publication. For this reason, there is a small degree of repetition of the key themes and background literature in the introductions, in order to place each chapter in its correct Chapter 1: Introduction 23

context. Chapter 2 briefly describes each of the species, both predator and prey, studied in this thesis. Chapters 3 and 4 examine the avoidance responses of native prey species to tiger quoll and red fox faecal odour in the field using the “scat at trap” technique

(sensu Powell and Banks 2004). Chapters 5 and 6 explore the spatial, foraging, and locomotory responses of native prey species to tiger quoll and red fox faecal odours in captive experiments. In Chapter 7, the volatile components of tiger quoll and red fox urine and faeces are analysed to determine if there are any common components prey species may be responding to. In Chapter 8, I investigate whether captive tiger quolls respond to the odours of native prey species, and Chapter 9 is a general discussion of all of the results, both in terms of how they relate to our general understanding of the role of odour in predator prey interactions, and their specific conservation implications. Chapter 1: Introduction 24

1.11 References

Abbott, D. H., Baines, D. A., Faulkes, C. G., Jennings, D. C., Ning, P. C. Y. K. and Tomlinson, A. J. (1990). A natural deer repellent: chemistry and behaviour. In Chemical signals in vertebrates 5. (edited by D. W. Macdonald, D. Muller- Schwarze and S. E. Natynczuk). Oxford University Press, Oxford.

Abbott, I. (2002). Origin and spread of the cat, Felis catus, on mainland Australia, with a discussion of the magnitude of its early impact on native fauna. Wildlife Research 29, 51-74.

Arnould, C., Malosse, C., Signoret, J.-P. and Descoins, C. (1998). Which chemical constituents from dog feces are involved in its food repellent effect in sheep? Journal of Chemical Ecology 24, 550-576.

Bailey, S., Bunyan, P. J. and Page, J. M. J. (1980). Variation in the levels of some components of the volatile fraction of urine from captive red foxes (Vulpes vulpes) and its relationship to the state of the animal. In Chemical Signals: Vertebrates and Aquatic Invertebrates. (edited by D. Muller-Schwarze and R. M. Silverstein). Plenum Press, New York.

Banks, P. B. (1998). Responses of Australian Bush Rats, Rattus fuscipes, to the odour of introduced Vulpes vulpes. Journal of Mammalogy 79, 1260-1264.

Banks, P. B., Hughes, N. K. and Rose, T. A. (2003). Do native Australian small mammals avoid faeces of domestic dogs? Responses of Rattus fuscipes and Antechinus stuartii. Australian Zoologist 32, 406-409.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2000). Nonlinearity in the predation risk of prey mobility. Proceedings of the Royal Society of London: Series B, Biological Sciences. 267, 1621-1625.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2002). Mobility decisions and the predation risks of reintroduction. Biological Conservation 103, 133-138.

Barreto, G. R. and Macdonald, D. W. (1999). The response of water voles, Arvicola terrestris, to the odours of predators. Animal Behaviour 57, 1107-1112.

Begon, M., Harper, J. L. and Townsend, C. R. (1996a). Ecology: Individuals, Populations and Communities. Blackwell Science, Cambridge.

Begon, M., Mortimer, M. and Thompson, D. J. (1996b). Population Ecology: A Unified Study of Animals and Plants. Blackwell Sciences, Cambridge.

Blumstein, D. T., Mari, M., Daniel, J. C., Ardon, J. G., Griffith, A. S. and Evans, C. S. (2002). Olfactory predator recognition: wallabies may have to learn to be wary. Animal Conservation 50, 143-198.

Bolbroe, T., Jeppesen, L. L. and Leirs, H. (2000). Behavioral response of field voles under mustelid predation risk in the laboratory: more than neophobia. Annales Zooloici Fennici 37, 169-178. Chapter 1: Introduction 25

Borowski, Z. (1998a). Influence of predator odour on the feeding behaviour of the root vole (Microtus oeconomus Pallas 1776). Canadian Journal of Zoology 76, 1791-1794.

Borowski, Z. (1998b). Influence of weasel (Mustela nivalis Linnaeus, 1766) odour on spatial behaviour of root voles (Microtus oeconomus Pallas, 1776). Canadian Journal of Zoology 76, 1799-1804.

Brahmachary, R. L. (1986). Ecology and chemistry of mammalian pheromones. Endeavour 10, 65-68.

Brinck, C., Erlinge, S. and Sandell, M. (1983). Anal sac secretion in Mustelids: A comparison. Journal of Chemical Ecology 9, 727-745.

Brown, J. S. (1988). Patch use as an indicator of habitat preference, predator risk, and competition. Behavioural Ecology and Socio-biology 22, 37-47.

Calder, C. J. and Gorman, M. L. (1991). The effects of Red Fox Vulpes vulpes faecal odours on the feeding behaviour of Orkney Voles Microtus arvalis. Journal of Zoology (London) 224, 599-606.

Carlsen, M., Lodal, J., Leirs, H. and Jensen, T. S. (1999). The effect of predation risk on body weight in the field vole, Microtus agrestis. Oikos 87, 277-285.

Christensen, P. E. S. (1980). A sad day for native fauna. Forest Focus 23, 3-12.

Corbet, G. B. and Wallis, R. L. (1977). Orkney/Guernsey vole Microtus arvalis. In The Handbook of British Mammals. (edited by G. B. Corbet and H. N. Southern). Blackwell Scientific Publications, Oxford.

Corbett, L. (1995). The Dingo in Australia and Asia. UNSW Press, Sydney.

Corbett, L. K. (1985). Morphological comparisons of Australian and Thai dingoes: a reappraisal of dingo status, distribution and ancestry. Proceedings of the Ecological Society of Australia 13, 277-291.

Crump, D. R. (1980). Thietanes and Dithiolanes from the anal gland of the stoat (Mustela erminea). Journal of Chemical Ecology 6, 341-347.

Curio, E. (1976). The Ethology of Predation. Springer-Verlag, Berlin.

Cushing, B. S. (1984). A Selective Preference by Least Weasels for Oestrous Versus Dioestrous Urine of Prairie Deer Mice. Animal Behaviour 32, 1263-1265.

Cushing, B. S. (1985). Estrous mice and vulnerability to weasel predation. Ecology 66, 1976-1978.

Dawkins, R. and Krebs, J. R. (1979). Arms races between and within species. Proceedings of the Royal Society of London B 205, 489-511.

Day, M. G. (1968). Food habits of British stoats (Mustela erminea) and weasels (Mustela nivalis). Journal of Zoology (London) 155, 485-497. Chapter 1: Introduction 26

Dickman, C. R. (1992). Predation and Habitat Shift in the House Mouse, Mus domesticus. Ecology 73, 313-322.

Dickman, C. R. (1996a). Impact of exotic generalist predators on the native fauna of Australia. Wildlife Biology 2, 185-195.

Dickman, C. R. (1996b). Overview of the impacts of feral cats on Australian native fauna. ANCA, Canberra.

Dickman, C. R. and Doncaster, C. P. (1984). Responses of Small Mammals to Red Fox (Vulpes vulpes) Odour. Journal of Zoology (London) 204, 521-531.

Dickman, C. R. and Doncaster, C. P. (1987). The ecology of small mammals in urban habitats. I. Populations in a patchy environment. Journal of Animal Ecology 56, 629-640.

Dickman, C. R., Pressey, R. L., Lim, L. and Parnaby, H. E. (1993). Mammals of particular conservation concern in the western division of New South Wales. Biological Conservation 65, 219-248.

Dielenberg, R. A., Carrive, P. and McGregor, I. S. (2001). The cardiovascular and behavioral response to cat odor in rats: unconditioned and conditioned effects. Brain Research 897, 228-237.

Dielenberg, R. A. and McGregor, I. S. (2001). Defensive behavior in rats towards predatory odors: a review. Neuroscience and Biobehavioral Reviews 25, 597- 609.

Doncaster, C. P., Dickman, C. R. and Macdonald, D. W. (1990). Feeding ecology of red foxes (Vulpes vulpes) in the city of Oxford, England. Journal of Mammalogy 71, 188-194.

Flannery, T. F. (1991). The mystery of the meganesian meat-eaters. Australian Natural History 23, 722-729.

Flannery, T. F. (1997). The Future Eaters. New Holland Publishers, Sydney.

Fuelling, O. and Halle, S. (2004). Breeding suppression in free-ranging grey-sided voles under the influence of predator odour. Oecologia 138, 151-159.

Godthelp, H. (1989). Pseudomys vandycki, A Tertiary Murid from Australia. Memoirs of the Queensland Museum 28, 171-173.

Gorman, M. L. (1980). Sweaty Mongooses and Other Smelly Carnivores. Symposium of the Zoological Society of London 45, 87-105.

Gorman, M. L. (1984). The response of prey to Stoat (Mustela erminea) scent. Journal of Zoology (London) 202, 419-423.

Gorman, M. L. and Trowbridge, B. J. (1989). The role of Odor in the Social Lives of Carnivores. In Carnivore Behavior, Ecology, and Evolution. (edited by J. L. Gittleman) pp. 57-88. Cornell University Press, New York. Chapter 1: Introduction 27

Gresser, S. (1996). Anti-predator Behaviour of the Common Brushtail Possum (Trichosurus vulpecula) at Burrendong Dam, N.S.W. Honours Thesis, University of Sydney, Sydney: 106.

Hecht, M. K. (1975). The morphology and relationships of the largest known terrestrial lizard, Megalania prisca Owen, from the Pleistocene of Australia. Proceedings of the Royal Society of Victoria 87, 239-250.

Herman, C. S. and Valone, T. J. (2000). The effect of mammalian predator scent on the foraging behaviour of Dipodomys merriami. Oikos 91, 139-145.

Jedrzejewski, W. and Jedrzejewska, B. (1990). Effect of a predator's visit on the spatial distribution of bank voles: experiments with weasels. Canadian Journal of Zoology 68, 660-666.

Jedrzejewski, W., Rychlik, W. and Jedrzejewska, B. (1993). Responses of bank voles to odours of seven species of predators: experimental data and their relevance to natural predator-vole relationships. Oikos 68, 251-257.

Jones, M. and Dayan, T. (2000). Foraging behaviour and microhabitat use by spiny mice Acomys cahirinus and A. russatus, in the presence of Blandford's fox (Vulpes vulpes) odor. Journal of Chemical Ecology 26, 455-469.

Jones, M. E., Smith, G. C. and Jones, S. M. (2004). Is anti-predator behaviour in Tasmanian eastern quolls (Dasyurus viverrinus) effective against introduced predators? Animal Conservation 7, 155-160.

Jonsson, P., Koskela, E. and Mappes, T. (2000). Does risk of predation by mammalian predators affect the spacing behaviour of rodents? Two large-scale experiments. Oecologia 122, 487-492.

Jorgenson, J. W., Novotny, M., Carmack, M., Copeland, G. B., Wilson, S. R., Katona, S. and Whitten, W. K. (1978). Chemical scent constituents in the urine of the red fox (Vulpes vulpes) during the winter season. Science 199, 796-798.

Kats, L. B. and Dill, L. M. (1998). The scent of death: Chemosensory assessment of predation risk by prey animals. Ecoscience 5, 361-394.

King, C. M. (1980). The weasel Mustela nivalis and its prey in an English woodland. Journal of Animal Ecology 49, 127-159.

King, C. M. (1990). The natural history of weasels & stoats. Cornell University Press, Ithaca.

Koivula, M. and Korpimäki, E. (2001). Do scent marks increase predation risk of microtine rodents? Oikos 95, 275-281.

Koskela, E. and Ylönen, H. (1995). Suppressed breeding in the field vole (Microtus agrestis): an adaptation to cyclically fluctuating predation risk. Behavioral Ecology 6, 311-315. Chapter 1: Introduction 28

Krebs, C. J. (1994). Ecology: the experimental analysis of distribution and abundance. HarperCollins College Publishers, New York.

Lever, R. J. A. (1959). The diet of the fox since myxomatosis. Journal of Animal Ecology 28, 359-375.

Lima, S. L. (1998). Nonlethal Effects in the Ecology of Predator-Prey Interactions: What are the ecological effects of anti-predator decision-making? Bioscience 48, 25-34.

Lima, S. L. (2002). Putting predators back into behavioral predator-prey interactions. Trends in Ecology and Evolution 17, 70-75.

Lima, S. L. and Dill, L. M. (1990). Behavioral decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology 68, 619-640.

Macdonald, D. W. (1980). Patterns of Scent Marking with Urine and Faeces Amongst Carnivores Communities. Symposium of the Zoological Society of London 45, 107-139.

Macdonald, D. W. (2001). The New Encyclopedia of Mammals. Oxford University Press, Oxford.

Macdonald, D. W. and Brown, R. E. (1985). Introduction: The pheromone concept in mammalian chemical communication. In Social odours in mammals. (edited by D. W. Macdonald and R. E. Brown). 1 pp. 1-36. Clarendon Press, Oxford.

Mappes, T., Koskela, E. and Ylönen, H. (1998). Breeding suppression in voles under predation risk of small mustelids: laboratory or methodological artifact. Oikos 82, 365-369.

Mappes, T. and Ylönen, H. (1997). Reproductive effort of female bank voles in a risky environment. Evolutionary Ecology 11, 591-598.

McDonald, R. A., Webbon, C. and Harris, S. (2000). The diet of stoats (Mustela erminea) and weasels (Mustela nivalis) in Great Britain. Journal of Zoology (London) 252, 363-371.

McGregor, I. S., Schram, L., Ambermoon, P. and Dielenberg, R. A. (2002). Not all 'predator odours' are equal: cat odour but not 2,3,5 trimethylthiazoline (TMT; fox odour) elicits specific defensive behaviour in rats. Behavioural Brain Research 129, 1-16.

McGregor, P. K. (1993). Signalling in territorial systems: a context for individual identification, ranging and eavesdropping. Philosophical Transactions of the Royal Society of London B 340, 237-244.

Montague, T. L., Pollock, D. C. and Wright, W. (1990). An examination of the Browsing animal problem in Australian eucalypt and pine plantations. Proceedings of Vertebrate Pest Conferences 14, 203-208. Chapter 1: Introduction 29

Moors, P. J. (1975). The food of weasels (Mustela nivalis) on farmland in north-east Scotland. Journal of Zoology 177, 455-461.

Morgan, D. R. and Woolhouse, A. D. (1997). Predator Odours as repellents to Brushtail Possums and Rabbits. In Repellents in Wildlife Management. (edited by J. R. Mason) pp. 241-252. National Wildlife Research Center, Ft. Collins.

Morton, S. R. (1990). The impact of European Settlement on the vertebrate animals of arid Australia: a conceptual model. Proceedings of the Ecological Society of Australia 16, 210-213.

Muller-Scwarze, D. (1972). Responses of young black-tailed deer to predator odors. Journal of Mammalogy 53, 393-394.

Nolte, D. L., Mason, J. R., Epple, G., Arnov, E. and Campbell, D. L. (1994). Why are predator urines aversive to prey. Journal of Chemical Ecology 20, 1505-1516.

Norrdahl, K. and Korpimäki, E. (1998). Does mobility or sex of voles affect risk of predation by mammalian predators? Ecology 79, 433-437.

Orrock, J. L., Danielson, B. J. and Brinkerhoff, R. J. (2004). Rodent foraging is affected by indirect, but not direct, cues of predation risk. Behavioural Ecology 15, 433- 437.

Österholm, H. (1964). The significance of distance receptors in the feeding behaviour of the fox (Vulpes vulpes L.). Acta Zoologica Fennica 106, 1-31.

Parsons, G. J. and Bondrup-Nielsen, S. (1996). Experimental analysis of behaviour of meadow voles (Microtus pennsylvanicus) to odours of the short-tailed weasel (Mustela erminea). Ecoscience 3, 63-69.

Powell, F. and Banks, P. B. (2004). Do house mice modify their foraging behaviour in response to predator odours and habitat? Animal Behaviour 67, 753-759.

Raymer, J., Wiesler, D., Novotny, M., Asa, C., Seal, U. S. and Mech, L. D. (1984). Volatile constituents of wolf (Canis lupus) urine as related to gender and season. Experientia 40, 707-709.

Raymer, J., Wiesler, D., Novotny, M., Asa, C., Seal, U. S. and Mech, L. D. (1986). Chemical scent constituents in urine of wolf (Canis lupus) and their dependence on reproductive hormones. Journal of Chemical Ecology 12, 297-314.

Regnier, F. E. and Goodwin, M. (1976). On the Chemical and Environmental Modulation of Pheromone Release from Vertebrate Scent Marks. In Chemical Signals in Vertebrates. (edited by D. Muller-Schwarze and M. M. Mozell) pp. 115-134. Plenum Press, New York.

Rich, T. H. and Hall, B. (1984). Rebuilding a giant lizard. In Vertebrate Zoogeography and Evolution in Australasia. (edited by A. M and G. Clayton). Hesperian Press, Sydney. Chapter 1: Introduction 30

Richards, D. F. (1977). Observations on the diet of the red fox (Vulpes vulpes) in South Devon. Journal of Zoology (London) 183, 495-504.

Richards, J. D. and Short, J. (2003). Reintroduction and establishment of the western barred bandicoot Perameles bougainville (Marsupialia: Peramelidae) at Shark Bay, Western Australia. Biological Conservation 109, 181-195.

Roberts, S. C., Gosling, L. M., Thornton, E. A. and McClung, J. (2001). Scent-marking by male mice under the risk of predation. Behavioral Ecology 12, 698-705.

Robinson, I. (1990). The effect of mink odour on rabbits and small mammals. In Chemical Signals in Vertebrates V. (edited by D. W. Macdonald). Oxford University Press, New York.

Rolls, E. C. (1969). They All Ran Wild, the Animals and Plants that Plague Australia. Angus and Robertson, Sydney.

Ronkainen, H. and Ylönen, H. (1994). Behaviour of cyclic bank voles under risk of mustelid predation: do females avoid copulation? Oecologia 97, 377-381.

Rosell, F. (2001). Effectiveness of predator odors as gray squirrel repellents. Canadian Journal of Zoology 79, 1719-1723.

Schultz, T. H., Flath, R. A., Stern, D. J., Mon, T. R., Teranishi, R., Kruse, S. M., Butler, B. and Howard, W. E. (1988). Coyote Estrous Urine Volatiles. Journal of Chemical Ecology 14, 701-712.

Schultz, T. H., Kruse, S. M. and Flath, R. A. (1985). Some volatile compoenents of female dog urine. Journal of Chemical Ecology 11, 169-175.

Short, J. (1998). The extinction of rat-kangaroos (Marsupialia: Potoroidae) in New South Wales, Australia. Biological Conservation 86, 365-377.

Short, J., Kinnear, J. E. and Robley, A. (2002). Surplus killing by introduced predators in Australia-evidence for ineffective anti-predator adaptations in native prey species? Biological Conservation 103, 283-301.

Sih, A. (1992). Prey Uncertainty and the Balancing of Antipredator and Feeding Needs. The American Naturalist 139, 1052-1069.

Smith, A. P. and Quin, D. G. (1996). Patterns and causes of extinction and decline in Australian conilurine rodents. Biological Conservation 77, 243-267.

Smith, J. N. M. (1974). The food searching behaviour of two European Thrushes. II: The adaptiveness of the search patterns. Behaviour 49, 1-61.

Sokolov, V. E., Albone, E. S., Flood, P. F., Heap, P. F., Kagan, M. Z., Vasilieva, V. S., Roznov, V. V. and Zinkevich, E. P. (1980). Secretion and secretory tissues of the anal sac of the mink, Mustela vison. Journal of Chemical Ecology 6, 805- 825. Chapter 1: Introduction 31

Southgate, R. I. (1990). Distribution and abundance of the Greater Bilby Macrotis lagotis Reid (Marsupialia: Peramelidae). In Bandicoots and Bilbies. (edited by J. H. Seebeck, P. R. Brown, R. L. Wallis and C. M. Kemper) pp. 293-302. Surrey Beatty and Sons, Sydney.

Stiling, P. (2002). Ecology: Theories and Applications. Prentice Hall, Upper Saddle River.

Stiling, P. D. (1992). Introductory Ecology. Prentice Hall, Englewood Cliff.

Stoddart, D. M. (1976a). Effect of Odour of Weasels (Mustela nivalis L.) on Trapped Samples of Their Prey. Oecologia 22, 439-441.

Stoddart, D. M. (1976b). Mammalian Odours and Pheromones. Edward Arnold Ltd, London.

Stoddart, D. M. (1982a). Demonstration of Olfactory Discrimination by the Short-tailed Vole, Microtus agrestis L. Animal Behaviour 30, 293-301.

Stoddart, D. M. (1982b). Does trap odour influence estimation of population size of the short-tailed vole, Microtus agrestis? Journal of Animal Ecology 51, 375-386.

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

Sullivan, T. P. and Crump, D. R. (1984). Influence of mustelid scent-gland compounds on suppression of feeding by snowshoe hares (Lepus americanus). Journal of Chemical Ecology 10, 1809-1821.

Sullivan, T. P. and Crump, D. R. (1986). Feeding responses of snowshoe hares (Lepus americanus) to volatile constituents of red fox (Vulpes vulpes) urine. Journal of Chemical Ecology 12, 729-739.

Sullivan, T. P., Nordstrom, L. O. and Sullivan, D. S. (1985a). Use of predator odors as repellents to reduce feeding damage by herbivores I. Snowshoe Hares (Lepus americanus). Journal of Chemical Ecology 11, 903-919.

Sullivan, T. P., Nordstrom, L. O. and Sullivan, D. S. (1985b). Use of predator odors as repellents to reduce feeding damage by herbivores II. Black-Tailed Deer (Odocoileus hemionus columbianus). Journal of Chemical Ecology 11, 921-935.

Thorsen, J. M., Morgan, R. A., Brown, J. S. and Norman, J. E. (1998). Direct and indirect cues of predatory risk and patch use by fox squirrels and thirteen-lined ground squirrels. Behavioural Ecology 9, 151-157.

Ward, J. F., Macdonald, D. W. and Doncaster, C. P. (1997). Responses of foraging hedgehogs to badger odour. Animal Behaviour 53, 709-720.

Watts, C. H. S. (1969). Distribution and Habits of the Rabbit Bandicoot. Transactions of the Royal Society of South Australia 93, 135-141.

Watts, C. H. S. and Aslin, H. J. (1981). The Rodents of Australia. Angus and Robertson Publishers, Sydney. Chapter 1: Introduction 32

Wells, M. C. (1978). Coyote senses in predation: environmental influences on their relative use. Behavioural Processes 3, 149-158.

Wells, M. C. and Lehner, P. N. (1978). The relative importance of the distance senses in coyote predatory behaviour. Animal Behaviour 26, 251-258.

Wilson, S. R., Carmack, M., Novotny, M., Jorgenson, J. W. and Whitten, W. K. (1978). delta3-Isopentenyl Methyl Sulfide. A New Terpenoid in the Scent Mark of the Red Fox (Vulpes vulpes). Journal of Organic Chemistry 43, 4675-4676.

Wolff, J. O. (2004). Scent marking by voles in response to predation risk: a field- laboratory validation. Behavioral Ecology 15, 286-289.

Wolff, J. O. and Davis-Born, R. (1997). Response of gray-tailed voles to odours of a mustelid predator: a field test. Oikos 79, 543-548.

Woolhouse, A. D. and Morgan, D. R. (1995). An evaluation of repellents to suppress browsing by possums. Journal of Chemical Ecology 21, 1571-1583.

Wroe, S. (2002). A review of terrestrial mammalian and reptilian carnivore ecology in Australian fossil faunas, and factors influencing their diversity: the myth of reptilian domination and its broader ramifications. Australian Journal of Zoology 50, 1-24.

Wroe, S. (2003). Australian marsupial carnivores: recent advances in palaeontology. In Predators with pouches: the biology of carnivorous marsupials. (edited by M. E. Jones, C. R. Dickman and M. Archer) pp. 102-124. CSIRO publishing, Collingwood.

Wroe, S., Myers, T., Seebacher, F., Kear, B., Gillespie, A., Crowther, M. and Salisbury, S. (2003). An alternative method of predicting body mass: The case of the Pleistocene marsupial lion. Paleobiology 29, 403-411.

Wroe, S., Myers, T. J., Wells, R. T. and Gillespie, A. (1999). Estimating the weight of the Pleistocene marsupial lion, Thylacoleo carnifex (Thlacoleonidae: Marsupialia): Implications for the ecomorphology of a marsupial super-predator and hypotheses of impoverishment of Australian marsupial carnivore faunas. Australian Journal of Zoology 47, 489-498.

Ylönen, H. (2001). Predator Odours and Behavioural Responses of Rodents: An Evolutionary Perspective. In Advances in Pest Management 2. (edited by H. J. Pelz, D. P. Cowan and C. J. Feare). Filander Verlag, Furth.

Ylönen, H. and Ronkainen, H. (1994). Breeding suppression in the bank vole as antipredatory adaptation in a predictable environment. Evolutionary Ecology 8, 658-666.

Ylönen, H., Sundell, J., Tiilikainen, R., Eccard, J. A. and Horne, T. (2003). Weasels' (Mustela nivalis nivalis) preference for olfactory cues of the vole (Clethrionomys glareolus). Ecology 84, 1447-1452. Chapter 2: Study Species 33

Chapter 2:

Study Species

2.1 Predator species

2.1.1 Australia’s mammalian predators

It has regularly been suggested over the last few decades that since the Miocene

Australia has had a depauperate mammalian carnivore assemblage (Hecht 1975; Rich and Hall 1984; Flannery 1991; Flannery 1997), and that this niche has instead been filled by reptiles (Hecht 1975; Flannery 1991; Flannery 1997). The re-evaluation of the sizes of several prehistoric predators, both mammalian and reptilian, has recently called this view into question (Wroe et al. 1999; Wroe 2002; Wroe 2003; Wroe et al. 2003).

Australia does have fewer species of mammalian carnivore than some other parts of the world, but this is in accordance with the principles of Island Biogeography theory

(MacArthur and Wilson 1967); that area and isolation of a landmass will affect species diversity due to interactions between speciation, extinction and immigration (Heaney

2000). Wroe (2003) also makes the point that species diversity is not the only factor in terms of the ecological significance of Australian mammalian carnivores, and that the ranges and abundance of species may be the more important factors.

The number of native mammalian carnivore species encountered by early British settlers was similar to the number of species they had left behind in Britain, which although much smaller than Australia, has only been isolated from the rest of the

Palaearctic region for the last 7000 years (Stokes 1961). In 1788 there were eight species of carnivorous mammal in Britain, in descending order of size: the Eurasian badger Meles meles, the otter Lutra lutra, the red fox Vulpes vulpes, the wild cat Felis Chapter 2: Study Species 34

silvestris, the pine marten Martes martes, the western polecat Mustela putorius, the stoat M. erminea and the weasel M. nivalis (Corbet and Southern 1977). The last few remaining populations of one other carnivore, the grey wolf Canis lupus lupus, had been hunted to extinction in Scotland and Ireland several decades earlier (Corbet and

Southern 1977). Only five of these species were found in Ireland, as weasels, polecats and wildcats have never been found there (Corbet and Southern 1977; Sleeman 1989).

However, all eight species were common throughout mainland Britain at this time, as the large range declines of the pine marten, polecat and wildcat, due to human persecution, did not occur until the latter half of the 19th century (Taylor 1946; Corbet and Southern 1977; Sleeman 1989).

In comparison, there were 11 carnivorous mammal species in Australia, again in descending order of size: the dingo Canis lupus dingo, the thylacine Thylacinus cynocephalus, the Tasmanian devil Sarcophilus harrisii, the tiger quoll Dasyurus maculatus, the western quoll D. geoffroii, the eastern quoll D. viverrinus, the water-rat

Hydromys chrysogaster, the northern quoll Dasyurus hallacatus, the brush-tailed phascogale Phascogale tapoatafa, the mulgara Dasycercus cristicaudata, and the kowari Dasyuroides byrnei (Strahan 1998). The water-rat has been included in this list as it fills a similar ecological niche as the otter. Similarly, I have included the brush- tailed phascogale, the mulgara and the kowari as these three dasyurids are larger than weasels, and actively prey upon small mammals and other small vertebrates, as well as invertebrates (Strahan 1998), just as I have included badgers in the British list despite their more omnivorous predilections (Macdonald and Barrett 1993). The highest concentration of Australian carnivorous mammals species would have been five species in both Tasmania where would have been found the thylacine, Tasmanian devil, tiger Chapter 2: Study Species 35

quoll, eastern quoll and water rat, and in the forested areas of the southeast coast of

Australia, from south-eastern South Australia, through Victoria and as far north as northern New South Wales, where would have been the dingo, tiger quoll, eastern quoll, water rat and brush-tailed phascogale (Strahan 1998).

When the British settled in Sydney quolls appear to have been particularly abundant, and thousands were poisoned, trapped and shot on a yearly basis throughout the early period of Australia’s colonial history, for their fur, for sport, and as a perceived pest

(Rolls 1969; Caughley 1980; Mansergh 1983). The added pressure of habitat clearance and an epidemic of an unknown disease at the beginning of the 20th century resulted in a drastic decline (Caughley 1980; Green and Scarborough 1990). Eastern quolls are now extinct on the mainland and found only in Tasmania (Jones and Rose 2001). Tiger quolls are extinct in South Australia and their distribution is thought to be disjunct throughout the rest of their range (Mansergh 1983). Since European settlement two other species of carnivore have been introduced and have become established and widespread throughout the Australian mainland, the red fox and the feral cat Felis catus

(Rolls 1969; Dickman 1996; Jones et al. 2003). In many areas where quolls are absent or scarce these introduced species have now taken over their role as medium-sized mammalian predators.

This thesis focuses on the tiger quoll and the red fox as both species use faeces in scent marking, defecating in prominent positions along trails, both natural and man-made

(Macdonald 1979; Macdonald 1985; Belcher 1995; Kruuk and Jarman 1995; Henry

1996; Triggs 1996; Burnett 2000). Chapter 2: Study Species 36

2.1.2 The Tiger Quoll

The tiger quoll (also known as the spotted-tailed or spot-tailed quoll) is a member of the marsupial family Dasyuridae, whose members are all insectivorous and/or carnivorous

(Strahan 1998). Listed as vulnerable to extinction nationally under the Australian

Commonwealth Environment Protection and Biodiversity Conservation Act (1999) and also on the schedules of the NSW Threatened Species Conservation Act 1995 (Lunney et al. 2000), it is the largest of the quoll species, and the largest marsupial predator on the mainland of Australia (Belcher 2004). There are two recognised subspecies: D. m. gracilis is found in northern Queensland, whilst D. m. maculatus is found from south- eastern Queensland through eastern New South Wales and Victoria and throughout

Tasmania (Figure 2.1), although the Tasmanian populations are genetically distinct from the mainland and probably deserve subspecific classification (Firestone et al.

1999). Tiger quolls are sexually size dimorphic as adult males are much larger than females, and this is more pronounced in the larger D. m. maculatus with an average male weight of 3.5 kg for males, versus 1.8 kg for females, than for the smaller D. m. gracilis with an average male weight of 1.6 kg, and an average female weight of 1.15 kg (Jones et al. 2001). The maximum weight for males is 7 kg and for females it is 4 kg

(Settle 1978). The tiger quoll was first recorded as the “Spotted Marten” by Governor

Phillip (Phillip 1789) and does superficially resemble the members of the Martes genus.

They have short legs, and a long tail, to about the same length as the body. Fur is rufous to dark brown, and it is the only quoll to have white spots on the tail as well as the body

(Jones et al. 2001). Chapter 2: Study Species 37

B. Russell

B. Russell B. Russell

Figure 2.1: The tiger quoll Dasyurus maculatus and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 38

Tiger quolls are considered semi-arboreal, the first toe on the hind foot (the hallux) is clawless and opposed to the rest, and the pads of the feet are striated, assumably to assist in climbing (Caughley 1980). In Tasmania, 11% of distance travelled was above the ground (Jones and Barmuta 2000), and in northeast Queensland, 17% of daytime radiolocations were in the rainforest canopy (Burnett 2000). This level of arboreal activity is reflected in their diet. Arboreal species accounted for 31.7% and 51.4% average percent biomass of male and female tiger quoll diets respectively in Tasmania

(Jones and Barmuta 2000), whilst in Victoria 39.4% by biomass are arboreal species

(Belcher 1995).

They consume a wide range of prey, and the exact composition of their diet varies from location to location, and over time (Settle 1978; Alexander 1980; Belcher 1995; Jones and Barmuta 1998). They have been observed to prey upon animals up to the size of small wallabies (Green and Scarborough 1990; Jones et al. 2001), but the most important prey items by biomass are medium-sized mammals (0.5-10 kg) such as possums, bandicoots and rabbits (Belcher 1995; Jones and Barmuta 1998; Burnett

2000). However, small mammals (<0.5kg) such as Antechinus and Rattus species are sometimes found in scats at a similar frequency (Alexander 1980; Belcher 1995;

Burnett 2000), and in New South Wales during Summer almost 100% of scats may contain invertebrate remains (Settle 1978; Alexander 1980).

Tiger quolls are found throughout a range of wooded habitats including wet scrub, coastal heathland, woodland, dry sclerophyll forest, wet sclerophyll forest, and rainforest (Edgar and Belcher 1998; Jones et al. 2001; Belcher 2004). They nest in a variety of natural hollows including rock crevices, caves, underground burrows, hollow Chapter 2: Study Species 39

logs and tree hollows, as well as the occasional manmade structure (Edgar and Belcher

1998; Burnett 2000; Jones et al. 2001). Whilst typically nocturnal, it is not unusual for them to be active during the daytime (Green and Scarborough 1990; Burnett 2000).

Although mean home range differs between sites, male home ranges may be over three times the size of female home ranges; Claridge et al. (2005) found mean male home range to be 992 ha and mean female home range to be 244 ha, and using a comparable method Belcher and Darrant (2004) found mean male average home range to be 1755 ha and mean female home range to be 496 ha. Throughout the year, male home ranges overlap with those of both sexes, whilst female home ranges do not overlap with other female home ranges unless they are mother/daughter (Belcher and Darrant 2004;

Claridge et al. 2005). Dispersal is male biased; males disperse upon maturity, whereas females remain within or close to their mother’s home range (Firestone 2003).

Tiger quolls scent mark using urine, faeces and cloacal dragging (Kruuk and Jarman

1995; Burnett 2000; Belcher and Darrant 2004). Communal latrines have been recorded for populations in New England (Kruuk and Jarman 1995), East Gippsland (Belcher

1995), Kosciuszko National Park (Claridge et al. 2004) and North East Queensland

(Burnett 2000), but have not been found for populations in Tasmania (Kruuk and

Jarman 1995). A latrine is defined as an aggregation of scats, where scats continue to be deposited over a period of time (Belcher 1995). In rocky areas latrines are usually located on top of large boulders with a flat horizontal surface (Kruuk and Jarman 1995;

Claridge et al. 2004), but may also be found in rocky creek beds, on rock shelves, at the base of cliffs, and along roads (Belcher 1995; Kruuk and Jarman 1995; Burnett 2000).

Claridge et al. (2004) found over 60 latrine sites in their study site in Kosciuszko

National Park, and Kruuk and Jarman (1995) found 29 latrine sites over one 500 m Chapter 2: Study Species 40

stretch, and 22 latrine sites over another 1 km stretch. Single scats may also be found at low density throughout the rest of the territory (Belcher 1995; Kruuk and Jarman 1995).

Tiger quolls are assumed to have a keen sense of smell which they utilise in predatory activities (Green and Scarborough 1990), and have been reported to be able to pick up the cross trails of rabbits Oryctolagus cuniculus (Fleay 1932), but their olfactory capabilities have yet to be systematically evaluated.

2.1.3 The Red Fox

The red fox is a member of the placental family Canidae, which include all the dog species, and are found naturally throughout Europe, Asia, North America and northern

Africa as far south as Sudan (Macdonald 2001). In Australia, red foxes are generally reddish-brown in colour, with a whitish chin, throat, chest and belly and a distinctive tail tag, which is usually white, but may be black or dark red (Coman 1998). Male foxes weigh between 4.7 and 8.3 kg, and females weigh between 4.0 and 6.8 kg (Coman

1998). They were introduced into Australia from Britain for the purposes of foxhunting

(Rolls 1969). Although foxes had been brought in to New South Wales and Victoria for hunts as early as 1845 or perhaps even earlier, they did not begin to breed and spread until two releases in the early 1870s at Ballarat and Point Cook in Victoria (Rolls

1969). Since that time they have spread throughout the whole of mainland Australia except for the tropical northern regions (Figure 2.2), and are found in almost all habitats, from desert and alpine regions to woodlands and forests as well as urban areas

(Coman 1998). Chapter 2: Study Species 41

A. Warn P.German

Figure 2.2: The red fox Vulpes vulpes and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 42

Although earlier attempts to introduce foxes to Tasmania failed (Saunders et al. 1995), more than 10 animals were deliberately released into Tasmania in 1998, and whilst few and far between, foxes have been consistently sighted since this time, and appear to be established in Tasmania, at least for the moment (Jones et al. 2004).

The early spread of foxes has been closely linked with the spread of rabbits, one of the more common dietary components of foxes in Europe, and the current broad distribution of foxes and rabbits is extremely similar (Jarman 1986; Saunders et al.

1995). Where both occur locally rabbits comprise the major component of fox diet

(Catling 1988; Saunders et al. 1995; Newsome et al. 1997), but foxes are characterised as being opportunistic predators (Lloyd 1977; Macdonald 1987; Macdonald and Barrett

1993; Saunders et al. 1995), and where rabbits are not readily available, foxes switch to medium sized marsupials such as possums and bandicoots and small mammals, mainly native rodents and Antechinus species, as their primary food source (Green and

Osbourne 1981; Triggs et al. 1984; Lunney et al. 1990; McKay 1994; Green 2003;

Mitchell and Banks 2005). Predation by the red fox is listed as a key threatening process under the Australian Commonwealth Environment Protection and Biodiversity

Conservation Act (1999) and the NSW Threatened Species Conservation Act 1995

(Mahon 2001).

As for tiger quolls, foxes are usually nocturnal, but may be seen during the day, especially in winter (Jarman 1986; Coman 1998). Dens may be either excavations of the burrows of other animals, or dug entirely by the fox, and foxes may also rest in thickets, hollow logs and under leaning trees (Jarman 1986; Coman 1998). Several dens may be found within the home range of an animal, which in Australia have been found Chapter 2: Study Species 43

to range from 63 to 700 ha (Coman et al. 1991; Phillips and Catling 1991; Meek and

Saunders 2000). Home ranges were occupied by either a single fox or a breeding pair

(Coman et al. 1991; Phillips and Catling 1991). Group living has been recorded in the alpine and subalpine areas of Kosciusko National Park (Bubela et al. 1998a), in an area where human refuse provides supplementary food especially during the winter months, maintaining the foxes at a density higher than could otherwise be supported (Bubela et al. 1998b).

Red foxes scent mark throughout their territories using both urine and faeces

(Macdonald 1979; Macdonald 1985). Faeces are deposited along trails, particularly at junctions or corners, in conspicuous places, such as grass tussocks, rocks, and clods of soil (Macdonald 1979; Macdonald 1985; Triggs 1996). Although foxes do not latrine as quolls do, it is not unusual to find accumulations of two or three scats in one place

(Macdonald 1985; Triggs 1996).

Henry (1980; 1996) described foxes lowering their head and sniffing in an attempt to locate prey, however when Österholm (1964) investigated the distance receptors of red foxes he found the sense of smell to be of less importance than commonly believed.

Foxes were able to olfactorially locate pieces of meat at a distance of up to two metres, but only if the fox was first aroused by an aural or visual cue (Österholm 1964). They were also able to detect buried pieces of meat at a depth of 10 cm (Österholm 1964).

Foxes clearly respond to conspecifics scent marks (Henry 1977; Macdonald 1979;

Macdonald 1985), but whether they respond to the scent marks of prey species has yet to be established. Chapter 2: Study Species 44

2.2 Prey species

The main prey species investigated in this thesis were the small marsupial: the brown antechinus Antechinus stuartii, the native rodents: the bush rat Rattus fuscipes and the swamp rat R. lutreolus, and the two bandicoot species: the long-nosed bandicoot

Perameles nasuta and the northern brown bandicoot Isoodon macrourus. Data were also collected in the field for the eastern chestnut mouse Pseudomys gracilicaudatus, the southern brown bandicoot I. obesulus and the common brushtail possum

Trichosurus vulpecula. Native rodents, antechinus, bandicoots and possums are the main groups of native mammals preyed upon by both tiger quolls and red foxes in

Australia.

2.2.1 The Brown Antechinus

The brown antechinus is a member of the marsupial family Dasyuridae, the same family to which quolls belong, but the brown antechinus is much smaller with males weighing on average 35 g, and females weighing 20 g (Braithwaite 1998a). Brown antechinus are nocturnal insectivores (Fox and Archer 1984; Braithwaite 1998a), ecologically most similar to shrews among placental species (Jarman 1986). With the recent description of the agile antechinus Antechinus agilis (Dickman et al. 1998), the range of the brown antechinus is now restricted to a broad coastal strip from Gympie in south-east Queensland to Kioloa in south-east New South Wales (Figure 2.3) (Sumner and Dickman 1998). Within this range it is common and abundant and occupies a variety of habitats including open forest, heathlands, woodland, and rainforest

(Braithwaite 1998a; Knight and Fox 2000). Chapter 2: Study Species 45

A. Poore

C. A. Henley A. Poore

Figure 2.3: The brown antechinus Antechinus stuartii and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 46

Brown antechinus nest in tree hollows and hollow logs (Braithwaite 1998a). Male and female antechinus nest communally in multiple nests after the previous year’s offspring are weaned (Lazenby-Cohen 1991). Antechinus are characterised by a highly synchronous two week breeding season during winter followed by complete mortality of all males in the population (Woolley 1966; Lee and Cockburn 1985). The timing of the two week breeding season varies throughout the species range (Dickman 1982) and is affected by photoperiod, although synchrony is based on olfactory cues from conspecific urine and faeces (Scott 1986). During the breeding season testosterone levels in males rise precipitously (Woolley 1966), resulting in tireless activity and increased aggression towards other males (Braithwaite 1974). During this time cloacal marking by dominant males has been observed following aggressive encounters

(Braithwaite 1974).

2.2.2 The Native Rodents

Australia’s rodent fauna is derived from two distinct waves to have arrived from New

Guinea (Strahan 1998). The ancestors of the “old endemics” entered Australia between

5 and 8 million years ago (Watts and Aslin 1981; Godthelp 1989; Watts and Baverstock

1995), and underwent an evolutionary radiation that has resulted in aquatic and arboreal species as well as terrestrial species, including a genus of hopping mice (Strahan 1998).

The ancestors of the “new endemics” entered Australia less than a million years ago

(Watts and Aslin 1981). All eight new endemic species belong to the Rattus genus and are relatively similar in form to other members of the genus (Strahan 1998). All of

Australia’s native rodents, both old endemic and new endemic, belong to the family

Muridae (Watts and Aslin 1981; Strahan 1998). Chapter 2: Study Species 47

Bush rats and swamp rats are both new endemic species of a similar size; bush rats weighing 40-225 g (Taylor and Calaby 1988a), and swamp rats weighing 56-200 g

(Taylor and Calaby 1988b; Monamy and Fox 2000). Swamp rats have darker fur, smaller ears, less protuberant eyes, and much darker feet than bush rats (Lunney

1998b). Their distributions are similar; both have isolated populations in northeast

Queensland, and then exist in a coastal band from southeast Queensland throughout

New South Wales and Victoria, to the southeast of South Australia, however the bush rat occurs in south west Western Australia (Figure 2.4) where the swamp rat does not, whilst the swamp rat is found in Tasmania (Figure 2.5) where the bush rat is absent

(Lunney 1998a; Lunney 1998b). Both species nest in excavated burrows (Taylor and

Calaby 1988a; Taylor and Calaby 1988b). Their social organisations are essentially the same (Lunney 1998a), after weaning both sexes establish discrete individual home ranges during winter, which are retained by females during the rest of the year, whilst males greatly expand their home ranges, which then overlap with many other males and females (Braithwaite and Lee 1979; Robinson 1987). Swamp rats are known to scent- mark with urine to a similar degree to black rats R. rattus and brown rats R. norvegicus

(Mallick 1992), and are attracted to conspecific scent marks (Mallick and Stoddart

1994). Similar work has not been conducted for bush rats, but they are thought to respond to the odour cues of conspecifics and competitors (Dickman and Woodside

1983).

Both species are omnivorous, consuming the stems and leaves of grasses and other monocots, as well as insects, fungus, fruits and seeds, although bush rats are more Chapter 2: Study Species 48

A. Poore

A. Poore A. C. Robinson

Figure 2.4: The bush rat Rattus fuscipes and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 49

N. Hughes

N. Hughes N. Hughes

Figure 2.5: The swamp rat Rattus lutreolus and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 50

insectivorous, and the swamp rat is more reliant on stems and leaves (Watts and

Braithwaite 1978; Cheal 1987; Luo and Fox 1996). There is a sharp delineation between the habitats occupied by these otherwise similar species, which appears to be facilitated by interference competition from the more dominant swamp rat (Maitz and

Dickman 2001). Bush rats are found in subalpine woodland, dry heath, eucalypt woodland and forest and rainforest, where there is a dense understorey of shrubs and/or ferns (Lunney 1998a). In Tasmania, where the bush rat is absent these habitats are also occupied by the swamp rat, where it is known as the velvet-furred rat, but on the mainland it is true to its more common name, being found predominantly in swamps and wet heath, creating its own tunnel system through the dense vegetation of grasses, sedges and reeds (Lunney 1998b).

The swamp rat shares this habitat preference with the new endemic species the eastern chestnut mouse (Fox 1998). Although described by Gould in 1845 from southeast

Queensland, this species was not recorded in New South Wales until the early 1970s

(Posamentier and Recher 1974; Mahony and Posamentier 1975). Their known current range extends along the east coast of Australia from Townsville in Queensland to

Brisabane Waters, just north of Sydney, in New South Wales (Figure 2.6) (Borshboom

1975; Mahony and Posamentier 1975; Fox 1998). The eastern chestnut mouse is listed as vulnerable to extinction in New South Wales on the schedules of the NSW

Threatened Species Conservation Act 1995 (Lunney et al. 2000). Although generally smaller than the swamp rat, weighing 45-118 g, it is otherwise superficially similar, the main distinguishing features being long grey-white hairs on the dorsal surface of the feet, lighter coloured feet, and a blunter more “mouse-like” snout (Fox 1998). Diet is similar as to swamp rats (Luo et al. 1994), although sympatric eastern Chapter 2: Study Species 51

D. Whitford

D. Whitford A. Poore

Figure 2.6: The eastern chestnut mouse Pseudomys gracilicaudatus and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 52

chestnut mice utilise different components as their staple to swamp rats (Luo and Fox

1996) and these staples change seasonally and successionally (Luo and Fox 1994). The two species are separated by time after fire and other disturbances, the optimum habitat for eastern chestnut mice is young regenerating wet heath, and they are competitively displaced by swamp rats as the vegetation matures (Higgs and Fox 1993). Eastern chestnut mice may nest in grass constructions above ground or in underground burrows

(Fox 1998). Little is known of their social organisation or use of olfaction.

2.2.3 The Bandicoots

Bandicoots belong to the marsupial family Peramelidae (Strahan 1998). Of the three species studied in this thesis the northern brown bandicoot is the largest, weighing between 500 and 3000 g, and is distributed in a coastal band from just north of the

Hawkesbury river (the northern boundary of Greater Sydney) in New South Wales, north to the tip of Cape York in Queensland, and then west through the Northern

Territory into the north of Western Australia (Figure 2.7) (Gordon 1998). The southern brown bandicoot weighs 400-1600 g, and is found throughout Tasmania, in southwest

Western Australia, in an isolated population in Cape York, and in a coastal strip from south of the Hawkesbury River throughout coastal Victoria into southeast South

Australia (Figure 2.8) (Braithwaite 1998a). It is listed as endangered on the NSW

Threatened Species Conservation Act 1995 (Lunney et al. 2000). Whilst these two

Isoodon species are essentially allopatric, the long-nosed bandicoot, weighing 520-1330 g (Scott et al. 1999), is sympatric with one of these two species throughout its range, which is a coastal strip from Northern Queensland, through New South Wales into eastern Victoria (Figure 2.9) (Stodart 1998). All three species are known Chapter 2: Study Species 53

J. Russell

A. C. Robinson A. Poore

Figure 2.7: The northern brown bandicoot Isoodon macrourus and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 54

J. Lochman K. J. Aslin

Figure 2.8: The southern brown bandicoot Isoodon obesulus and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 55

N. Hughes

N. Hughes N. Hughes

Figure 2.9: The long-nosed bandicoot Perameles nasuta and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 56

from various habitats including grassland, eucalypt woodland and forest (Braithwaite

1998b; Gordon 1998; Stodart 1998). However long-nosed bandicoots would appear to utilise areas with both dense vegetation for nesting and open areas for foraging

(Chambers and Dickman 2002), whilst the two Isoodon species are found far less often than long-nosed bandicoots in open environments (Gordon 1974; Claridge et al. 1991).

Like all bandicoot species, long-nosed bandicoots, northern brown bandicoots and southern brown bandicoots are omnivorous, consuming plant material including fruits, seeds and tubers as well as fungus, small vertebrates and invertebrates (Quin 1988; Gott

1996; Scott et al. 1999). The conical excavations which typify these species, much to the disgust of some suburban lawn owners, and the delight of others, are evidence of their subterranean search for food, but food may also be taken from the surface

(Braithwaite 1998b; Gordon 1998; Stodart 1998). All three species build a similar nest consisting of a shallow depression covered by grass, sticks and leaves, that is generally indistinguishable from the surrounding leaf litter (Stodart 1966; Gordon 1974; Lobert

1990) and each animal may have multiple nests active within its home range (Gordon

1974). Although solitary, in all three species home ranges overlap between and within the sexes, and male territories tend to be larger than female territories (Gordon 1974;

Broughton and Dickman 1991; Scott et al. 1999). A subauricular gland is present in all three species (Stoddart 1980), which is used for marking the ground and vegetation during aggressive encounters, mainly by males during the breeding season (Gordon

1998; Scott et al. 1999). Chapter 2: Study Species 57

2.2.4 The Common Brushtail Possum

The most common member of the marsupial family Phalangeridae, the common brushtail possum is a medium sized animal weighing between 1200 and 4500 g (How and Kerle 1998; Kerle 2001). It is a folivore, with the bulk of the diet being leaves, supplemented by fruits and flowers, grasses and occasionally insects and small vertebrates (How and Kerle 1998; Kerle 2001). The actual composition of the diet varies widely between locations, both in terms of the relative contributions of the different components and the species consumed (Fitzgerald 1984; Statham 1984; How and Hillcox 2000). It is found in a wide variety of wooded habitats from rainforest, through open forest and heath to open woodland (Kerle 1984), and is found in all states and territories (How and Kerle 1998), being absent only from Australia’s arid centre away from major waterways (Kerle et al. 1992) (Figure 2.10). Although their home ranges may overlap, especially in areas of high density, common brushtail possums are generally solitary and scent mark their territories with excretions from glands on the chin, chest and anus (Kerle 2001). Chapter 2: Study Species 58

A. Poore

J. Russell A. Poore

Figure 2.10: The common brushtail possum Trichosurus vulpecula and its distribution in Australia (after Strahan 1994). Chapter 2: Study Species 59

2.3 References

Alexander, D. (1980). Diet of the Tiger Quoll (Dasyurus maculatus Kerr) in an area of New England National Park, Northeastern New South Wales. Honours thesis, University of New England, Armidale.

Belcher, C. (2004). The largest surviving marsupial carnivore on mainland Australia: the Tiger or Spotted-tailed Quoll Dasyurus maculatus, a nationally threatened, forest dependent species. In Conservation of Australia's Forest Fauna. (edited by D. Lunney). Royal Zoological Society of New South Wales, Sydney.

Belcher, C. A. (1995). Diet of the tiger quoll (Dasyurus maculatus) in East Gippsland, Victoria. Wildlife Research 22, 341-357.

Belcher, C. A. and Darrant, J. P. (2004). Home range and spatial organization of the marsupial carnivore, Dasyurus maculatus maculatus (Marsupialia) in south- eastern Australia. Journal of Zoology (London) 262, 271-280.

Borshboom, A. C. (1975). Pseudomys gracilicaudatus (Gould): Range Extension and Notes on a Little-known Australian Murid Rodent. Australian Journal of Wildlife Research 2, 81-84.

Braithwaite, R. W. (1974). Behavioural Changes associated with the Population Cycle of Antechinus stuartii (marsupialia). Australian Journal of Zoology 22, 45-62.

Braithwaite, R. W. (1998a). Brown Antechinus Antechinus stuartii. In The Mammals of Australia. (edited by R. Strahan) pp. 94-97. New Holland Publishers, Sydney.

Braithwaite, R. W. (1998b). Southern Brown Bandicoot. In The Mammals of Australia. (edited by R. Strahan) pp. 176-177. New Holland Publishers, Sydney.

Braithwaite, R. W. and Lee, A. K. (1979). The Ecology of Rattus lutreolus I. A Victorian Heathland Population. Australian Wildlife Research 6, 173-189.

Broughton, S. K. and Dickman, C. R. (1991). The effect of supplementary food on home range of the southern brown bandicoot, Isoodon obesulus (Marsupialia: Peramelidae). Australian Journal of Ecology 16, 71-78.

Bubela, T., Bartell, R. and Muller, W. (1998a). Factors affecting the trappability of red foxes in Kosciusko National Park. Wildlife Research 25, 199-208.

Bubela, T. M., Dickman, C. R. and Newsome, A. E. (1998b). Diet and winter foraging behaviour of the red fox (Vulpes vulpes) in alpine and subalpine New South Wales. Australian Mammalogy 20, 321-330.

Burnett, S. E. (2000). Ecology and Conservation Status of the northern Spot-tailed Quoll, Dasyurus maculatus: with reference to the Future of Australia's Marsupial Carnivores. Ph.D. dissertation, James Cook University of Northern Queensland, Townsville: 292. Chapter 2: Study Species 60

Catling, P. C. (1988). Similarities and Contrasts in the Diets of Foxes, Vulpes vulpes, and Cats, Felis catus, Relative to Fluctuating Prey Populations and Drought. Australian Wildlife Research 15, 307-317.

Caughley, J. (1980). Native Quolls and Tiger Quolls. In Endangered Animals of NSW. (edited by C. Haigh) pp. 45-48. NSW National Parks and Wildlife Service, Sydney.

Chambers, L. K. and Dickman, C. R. (2002). Habitat selection of the long-nosed bandicoot, Perameles nasuta (Mammalia, Peramelidae), in a patchy urban environment. Austral Ecology 27, 334-342.

Cheal, D. C. (1987). The Diets and Dietary Preferences of Rattus fuscipes and Rattus lutreolus at Walkerville in Victoria. Australian Wildlife Research 14, 35-44.

Claridge, A. W., McNee, A., Tanton, M. T. and Davey, S. M. (1991). Ecology of bandicoots in undisturbed forest adjacent to recently felled logging coupes: a case study from the Eden Woodchip Agreement Area. In Conservation of Australia's Forest Fauna. (edited by D. Lunney). Royal Zoological Society of New South Wales, Mosman.

Claridge, A. W., Mifsud, G., Dawson, J. and Saxon, M. J. (2004). Use of infrared digital cameras to investigate the behaviour of cryptic species. Wildlife Research 31, 645-650.

Claridge, A. W., Paull, D., Dawson, J., Mifsud, G., Murray, A. J., Poore, R. and Saxon, M. J. (2005). Home range of the spotted-tailed quoll (Dasyurus maculatus), a marsupial carnivore, a in a rainshadow woodland. Wildlife Research 32, 7-14.

Coman, B. J. (1998). Fox Vulpes vulpes. In The Mammals of Australia. (edited by R. Strahan) pp. 698-699. New Holland Publishers, Sydney.

Coman, B. J., Robinson, J. and Beaumont, C. (1991). Home Range, Dispersal and Density of Red Foxes (Vulpes vulpes L.) in Central Victoria. Wildlife Research 18, 215-23.

Corbet, G. B. and Southern, H. N. (1977). The Handbook of British Mammals. Blackwell Scientific Publications, Oxford.

Dickman, C. R. (1982). Some ecological aspects of seasonal breeding in Antechinus (Dasyuridae: Marsupialia). In Carnivorous Marsupials. (edited by M. Archer) pp. 139-150. Royal Zoological Society of New South Wales, Sydney.

Dickman, C. R. (1996). Impact of exotic generalist predators on the native fauna of Australia. Wildlife Biology 2, 185-195.

Dickman, C. R., Parnaby, H. E., Crowther, M. S. and King, D. H. (1998). Antechinus agilis (Marsupialia: Dasyuridae), a new species from the A. stuartii complex in south-eastern Australia. Australian Journal of Zoology 46, 1-26. Chapter 2: Study Species 61

Dickman, C. R. and Woodside, D. P. (1983). A test of a competition model with reference to three species of small mammals in south-eastern Australia. Oecologia 60, 127-134.

Edgar, R. and Belcher, C. A. (1998). Spotted Tailed quoll Dasyurus maculatus. In The Mammals of Australia. (edited by R. Strahan). New Holland Publishers, Sydney.

Firestone, K. B. (2003). The application of genetic research to conservation management in carnivorous marsupials with special emphasis on dasyurids. In Predators with pouches: the biology of carnivorous marsupials. (edited by M. E. Jones, C. R. Dickman and M. Archer) pp. 475-485. CSIRO Publishing, Collingwood.

Firestone, K. B., Elphinstone, M. S., Sherwin, W. B. and Houlden, B. A. (1999). Phylogeographical population structure of tiger quolls Dasyurus maculatus (Dasyuridae: Marsupialia), an endangered carnivorous marsupial. Molecular Ecology 8, 1613-1625.

Fitzgerald, A. (1984). Diet of the possum (Trichosurus vulpecula) in three Tasmanian forest types and its relevance to the diet of possums in New Zealand forests. In Possums and Gliders. (edited by A. Smith and I. Hume) pp. 137-143. Surrey Beatty and Sons, Sydney.

Flannery, T. F. (1991). The mystery of the meganesian meat-eaters. Australian Natural History 23, 722-729.

Flannery, T. F. (1997). The Future Eaters. New Holland Publishers, Sydney.

Fleay, D. (1932). The rare dasyures (native cats). Victorian Naturalist 49, 63-69.

Fox, B. J. (1998). Eastern Chestnut mouse Pseudomys gracilicaudatus. In The Mammals of Australia. (edited by R. Strahan) pp. 601-603. New Holland Publishers, Sydney.

Fox, B. J. and Archer, E. (1984). The Diets of Sminthopsis murina and Antechinus stuartii (Marsupialia: Dasyuridae) in Sympatry. Australian Wildlife Research 11, 235-48.

Godthelp, H. (1989). Pseudomys vandycki, A Tertiary Murid from Australia. Memoirs of the Queensland Museum 28, 171-173.

Gordon, G. (1974). Movements and Activity of the Shortnosed Bandicoot Isoodon macrourus Gould (Marsupialia). Mammalia 38, 405-431.

Gordon, G. (1998). Northern Brown Bandicoot Isoodon macrourus. In The Mammals of Australia. (edited by R. Strahan) pp. 174-175. New Holland Publishers, Sydney.

Gott, M. (1996). Ecology of the Northern Brown Bandicoot, Isoodon macrourus: reproduction and resource use in a Heathland population. Ph.D. dissertation, University of New South Wales, Sydney. Chapter 2: Study Species 62

Green, K. (2003). Altitudinal and temporal differences in the food of foxes (Vulpes vulpes) at alpine and subalpine altitudes in the Snowy Mountains. Wildlife Research 30, 245-253.

Green, K. and Osbourne, W. S. (1981). The diet of foxes Vulpes vulpes (L), in relation to abundance of prey above the winter snowline in New South Wales. Australian Wildlife Research 87, 239-250.

Green, R. H. and Scarborough, T. J. (1990). The spotted-tailed quoll Dasyurus maculatus (Dasyuridae, Marsupialia) in Tasmania. The Tasmanian Naturalist 100, 1-15.

Heaney, L. R. (2000). Dynamic disequilibrium: a long-term, large-scale perspective on the equilibrium model of island biogeography. Global Ecology and Biogeography 9, 59-74.

Hecht, M. K. (1975). The morphology and relationships of the largest known terrestrial lizard, Megalania prisca Owen, from the Pleistocene of Australia. Proceedings of the Royal Society of Victoria 87, 239-250.

Henry, J. D. (1977). The use of urine marking in the scavenging behaviour of the red fox (Vulpes vulpes). Behaviour 61, 82-105.

Henry, J. D. (1980). Fox Hunting. Natural History 89, 60-69.

Henry, J. D. (1996). Red Fox: The Catlike Canine. Smithsonian Institute Press, Washington D.C.

Higgs, P. and Fox, B. J. (1993). Interspecific competition: A mechanism for rodent succession after fire in wet heathland. Australian Journal of Ecology 18, 193- 201.

How, R. A. and Hillcox, S. J. (2000). Brushtail possum, Trichosurus vulpecula, populations in south-western Australia: demography, diet and conservation status. Wildlife Research 27, 81-89.

How, R. A. and Kerle, J. A. (1998). Common Brushtail Possum Trichosurus vulpecula. In The Mammals of Australia. (edited by R. Strahan). New Holland Publishers, Sydney.

Jarman, P. (1986). The red fox - an exotic, large predator. In The Ecology of Exotic Animals and Plants: some Australian Case Histories. (edited by R. L. Kitching) pp. 43-61. John Wiley and Sons, Brisbane.

Jones, M. E. and Barmuta, L. A. (1998). Diet overlap and abundance of sympatric dasyurid carnivores: a hypothesis of competition? Journal of Animal Ecology 81, 434-447.

Jones, M. E. and Barmuta, L. A. (2000). Niche differentiation among sympatric Australian dasyurid carnivores. Journal of Mammalogy 81, 434-447. Chapter 2: Study Species 63

Jones, M. E., Oakwood, M., Belcher, C. A., Morris, K., Murray, A. J., Woolley, P. A., Firestone, K. B., Johnson, B. and Burnett, S. (2003). Carnivore concerns: Problems, issues and solutions for conserving Australasia's marsupial carnivores. In Predators with pouches: The biology of carnivorous marsupials. (edited by M. E. Jones, C. R. Dickman and M. Archer) pp. 422-434. CSIRO publishing, Collingwood.

Jones, M. E. and Rose, R. K. (2001). Dasyurus viverrinus. Mammalian Species 677, 1- 9.

Jones, M. E., Rose, R. K. and Burnett, S. (2001). Dasyurus maculatus. Mammalian Species 676, 1-9.

Jones, M. E., Smith, G. C. and Jones, S. M. (2004). Is anti-predator behaviour in Tasmanian eastern quolls (Dasyurus viverrinus) effective against introduced predators? Animal Conservation 7, 155-160.

Kerle, J. A. (1984). Variation in the ecology of Trichosurus: its adaptive significance. In Possums and Gliders. (edited by A. P. Smith and I. D. Hume) pp. 115-128. Surrey Beatty and Sons, Sydney.

Kerle, J. A. (2001). Possums: the brushtails, ringtails and greater glider. University of New South Wales Press, Sydney.

Kerle, J. A., Foulkes, J. N., Kimber, R. G. and Papenfus, D. (1992). The decline of the Brushtail Possum, Trichosurus vulpecula (Kerr 1798), in Arid Australia. Rangeland Journal 14, 107-127.

Knight, E. H. and Fox, B. J. (2000). Does habitat structure mediate the effects of forest fragmentation and human-induced disturbance on the abundance of Antechinus stuartii. Australian Journal of Zoology 48, 577-595.

Kruuk, H. and Jarman, P. J. (1995). Latrine use by the Spotted-tailed quoll (Dasyurus maculatus: Dasyuridae, Marsupialia) in its natural habitat. Journal of Zoology (London) 236, 345-349.

Lazenby-Cohen, K. A. (1991). Communal Nesting in Antechinus stuartii (Marsupialia: Dasyuridae). Australian Journal of Zoology 39, 272-283.

Lee, A. K. and Cockburn, A. (1985). Evolutionary ecology of marsupials. Cambridge University Press, Cambridge.

Lloyd, H. G. (1977). Fox Vulpes vulpes. In The Handbook of British Mammals. (edited by G. B. Corbet and H. N. Southern) pp. 311-320. Blackwell Scientific Publications, London.

Lobert, B. (1990). Home range and activity period of the Southern Brown Bandicoot (Isoodon obesulus) in a Victorian heathland. In Bandicoots and Bilbies. (edited by J. H. Seebeck, P. R. Brown, R. L. Wallis and C. M. Kemper) pp. 319-25. Surrey Beatty and Sons, Sydney. Chapter 2: Study Species 64

Lunney, D. (1998a). Bush Rat Rattus fuscipes. In The Mammals of Australia. (edited by R. Strahan) pp. 651-653. New Holland Publishers, Sydney.

Lunney, D. (1998b). Swamp Rat Rattus lutreolus. In The Mammals of Australia. (edited by R. Strahan) pp. 655-657. New Holland Publishers, Sydney.

Lunney, D., Curtin, A. L., Ayers, D., Cogger, H. G., Dickman, C. R., Maitz, W., Law, B. and Fisher, D. (2000). The threatened and non-threatened native vertebrate fauna of New South Wales: status and ecological attributes. NPWS, Hurtsville.

Lunney, D., Triggs, B., Eby, P. and Ashby, E. (1990). Analysis of scats of dogs Canis familiaris and foxes Vulpes vulpes (Canidae: Carnivora) in coastal forests near Bega, New South Wales. Australian Wildlife Research 17, 61-68.

Luo, J. and Fox, B. J. (1994). Diet of the Eastern Chestnut Mouse Pseudomys gracilicaudatus). II. Seasonal and Successional Patterns. Wildlife Research 21, 419-31.

Luo, J. and Fox, B. J. (1996). Seasonal and successional dietary shifts of two sympatric rodents in coastal heathland: A possible mechanism for coexistence. Australian Journal of Ecology 21, 121-132.

Luo, J., Fox, B. J. and Jeffreys, E. (1994). Diet of the Eastern Chestnut Mouse (Pseudomys gracilicaudatus). I. Composition, Diversity and Individual Variation. Wildlife Research 21, 401-17.

MacArthur, R. H. and Wilson, E. O. (1967). The theory of Island Biogeography. Princeton University Press, Princeton.

Macdonald, D. W. (1979). Some Observations and Field Experiments on the Urine Marking Behaviour of the Red Fox Vulpes vulpes L. Zeitshrift fuer Tierpsychologie 51, 1-22.

Macdonald, D. W. (1985). The carnivores: order Carnivora. In Social odours in mammals. (edited by R. E. Brown and D. W. Macdonald). 2 pp. 619-722. Clarendon Press, Oxford.

Macdonald, D. W. (1987). Running with the fox. Facts on File Publications, New York.

Macdonald, D. W. (2001). Foxes. In The New Encyclopedia of Mammals. (edited by D. W. Macdonald) pp. 54-61. Oxford University Press, Oxford.

Macdonald, D. W. and Barrett, P. (1993). Collins Field Guide: Mammals of Britain and Europe. Harper Collins, London.

Mahon, P. (2001). Draft Threat Abatement Plan for Predation by the Red Fox (Vulpes vulpes). NSW National Parks and Wildlife Service, Hurstville.

Mahony, J. A. and Posamentier, H. (1975). The occurence of the native rodent Pseudomys gracilicaudatus (Gould 1845) [Rodentia: Muridae] in New South Wales. Australian Mammalogy 1, 333-346. Chapter 2: Study Species 65

Maitz, W. E. and Dickman, C. R. (2001). Competition and habitat use in native Australian Rattus: is competition intense, or important? Oecologia 128, 526- 538.

Mallick, S. A. (1992). Urine-marking in Three Species of Rattus. Wildlife Research 19, 89-93.

Mallick, S. A. and Stoddart, D. M. (1994). Scent Marking and Perception in the Australian Swamp Rat Rattus lutreolus. Polish Ecological Studies 20, 393-402.

Mansergh, I. (1983). The Status and Abundance of Dasyurus maculatus (Tiger Quoll) in Australia, with particular reference to Victoria. The Australian Zoologist 21, 109-122.

McKay, G. M. (1994). Effects of introduced predators on biodiversity remnant vegetation. Australian Nature Conservation Agency, Canberra.

Meek, P. D. and Saunders, G. (2000). Home range and movement of foxes (Vulpes vulpes) in coastal New South Wales, Australia. Wildlife Research 27, 663-668.

Mitchell, B. D. and Banks, P. B. (2005). Do wild dogs exclude foxes? Evidence for competition from dietary and spatial overlaps. Austral Ecology 30, 581-591.

Monamy, V. and Fox, B. J. (2000). Small mammal succession is determined by vegetation density rather than time elapsed since disturbance. Austral Ecology 25, 580-587.

Newsome, A., Pech, R., Smith, R., Banks, P. and Dickman, C. (1997). Potential Impacts on Australian native fauna of rabbit calicivirus disease. Biodiversity Group, Environment Australia, Canberra.

Österholm, H. (1964). The significance of distance receptors in the feeding behaviour of the fox (Vulpes vulpes L.). Acta Zoologica Fennica 106, 1-31.

Phillip, A. (1789). The voyage of Governor Phillip to Botany Bay. J. Stockdale, London.

Phillips, M. and Catling, P. C. (1991). Home Range and Activity Patterns of Red Foxes in Nadgee Nature Reserve. Wildlife Research 18, 677-86.

Posamentier, H. and Recher, H. F. (1974). The Status of Pseudomys novaehollandiae (the New Holland Mouse). Australian Zoologist 18, 66-71.

Quin, D. G. (1988). Observations on the diet of the Southern Brown Bandicoot, Isoodon obesulus (Marsupialia: Peramelidae) in Southern Tasmania. Australian Mammalogy 11, 15-25.

Rich, T. H. and Hall, B. (1984). Rebuilding a giant lizard. In Vertebrate Zoogeography and Evolution in Australasia. (edited by A. M and G. Clayton). Hesperian Press, Sydney. Chapter 2: Study Species 66

Robinson, A. C. (1987). The ecology of the Bush Rat, Rattus fuscipes (Rodentia: Muridae), in Sherbrooke Forest, Victoria. Australian Mammalogy 11, 35-49.

Rolls, E. C. (1969). They All Ran Wild, the Animals and Plants that Plague Australia. Angus and Robertson, Sydney.

Saunders, G., Coman, B., Kinnear, J. and Braysher, M. (1995). Managing Vertebrate Pests: Foxes. Australian Government Publishing Service, Canberra.

Scott, L. K., Hume, I. D. and Dickman, C. R. (1999). Ecology and population biology of Long-nosed Bandicoots (Perameles nasuta) at North Head, Sydney Harbour National Park. Wildlife Research 26, 805-821.

Scott, M. P. (1986). The timing and synchrony of seasonal breeding in the marsupial, Antechinus stuartii: interaction of environmental and social cues. Journal of Mammalogy 67, 551-560.

Settle, G. A. (1978). The quiddity of tiger quolls. Australian Natural History 19, 164- 169.

Sleeman, P. (1989). Stoats & weasels: polecats & martens. Whittet Books, London.

Statham, H. L. (1984). The diet of Trichosurus vulpecula (Kerr) in four Tasmanian forest locations. In Possums and Gliders. (edited by A. Smith and I. Hume) pp. 213-219. Surrey Beatty and Sons, Sydney.

Stodart, E. (1966). Management and Behaviour of breeding groups of the marsupial Perameles nasuta Geoffroy in captivity. Australian Journal of Zoology 14, 611- 623.

Stodart, E. (1998). Long-nosed Bandicoot Perameles nasuta. In The Mammals of Australia. (edited by R. Strahan) pp. 184-185. New Holland Publishers, Sydney.

Stoddart, D. M. (1980). Observations of the Structure and Function of Cephalic Skin Glands in Bandicoots (Marsupialia: Peramelidae). Australian Journal of Zoology 28, 33-41.

Stokes, W. L. (1961). Essentials of Earth History: An Introduction to Historical Geography. Prentice Hall, New Jersey.

Strahan, R. (1994). Australian Mammals: the multimedia experience. Frenchs Forest, Webster Publishing.

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

Sumner, J. and Dickman, C. R. (1998). Distribution and identity of species in the Antechinus stuartii-A. flavipes group (Marsupialia: Dasyuridae) in south-eastern Australia. Australian Journal of Zoology 46, 27-41.

Taylor, J. M. and Calaby, J. H. (1988a). Rattus fuscipes. Mammalian Species 298, 1-8.

Taylor, J. M. and Calaby, J. H. (1988b). Rattus lutreolus. Mammalian Species 299, 1-7. Chapter 2: Study Species 67

Taylor, W. L. (1946). The wild cat (Felis sylvestris) in Great Britain. Journal of Animal Ecology 15, 130-133.

Triggs, B. (1996). Tracks, scats and other traces; A field guide to Australian mammals. Oxford University Press, South Melbourne.

Triggs, B., Brunner, H. and Cullen, J. M. (1984). The food of fox, dog and cat in Croajingalong National Park, south-eastern Victoria. Australian Wildlife Research 11, 491-499.

Watts, C. H. S. and Aslin, H. J. (1981). The Rodents of Australia. Angus and Robertson Publishers, Sydney.

Watts, C. H. S. and Baverstock, P. R. (1995). Evolution in the Murinae (Rodentia) assessed by Microcomplement Fixation of Albumin. Australian Journal of Zoology 43, 105-118.

Watts, C. H. S. and Braithwaite, R. W. (1978). The Diet of Rattus lutreolus and Five other rodents in Southern Victoria. Australian Wildlife Research 5, 47-57.

Woolley, P. (1966). Reproduction in Antechinus spp. and other dasyurid marsupials. Symposium of the Zoological Society of London 15, 281-94.

Wroe, S. (2002). A review of terrestrial mammalian and reptilian carnivore ecology in Australian fossil faunas, and factors influencing their diversity: the myth of reptilian domination and its broader ramifications. Australian Journal of Zoology 50, 1-24.

Wroe, S. (2003). Australian marsupial carnivores: recent advances in palaeontology. In Predators with pouches: the biology of carnivorous marsupials. (edited by M. E. Jones, C. R. Dickman and M. Archer) pp. 102-124. CSIRO publishing, Collingwood.

Wroe, S., Myers, T., Seebacher, F., Kear, B., Gillespie, A., Crowther, M. and Salisbury, S. (2003). An alternative method of predicting body mass: The case of the Pleistocene marsupial lion. Paleobiology 29, 403-411.

Wroe, S., Myers, T. J., Wells, R. T. and Gillespie, A. (1999). Estimating the weight of the Pleistocene marsupial lion, Thylacoleo carnifex (Thlacoleonidae: Marsupialia): Implications for the ecomorphology of a marsupial super-predator and hypotheses of impoverishment of Australian marsupial carnivore faunas. Australian Journal of Zoology 47, 489-498. Chapter 3: Responses of CWR marsupials to predator odours 68

Chapter 3:

Responses of four Critical Weight Range (CWR) marsupials to the odours of native and introduced predators

Published as: Russell, B. G. and Banks, P. B. (2005). Responses of four

Critical Weight Range (CWR) marsupials to the odours of native and introduced predators. Australian Zoologist 33: 217-222

Apart from minor formatting changes, this chapter appears as it does in Australian

Zoologist, hence the use of we instead of I. I designed and conducted the experiment, analysed the data, and wrote the paper in consultation with my supervisor Dr Peter

Banks.

3.1 Abstract

In Australia, many critical weight range (CWR) species are threatened by predation from the introduced red fox Vulpes vulpes. Understanding how these prey species respond to native predators such as the tiger quoll Dasyurus maculatus, and comparing their responses to foxes is important in understanding why fox predation is such a problem. Many northern hemisphere mammalian species have developed responses to the odours of the main species which prey upon them. The situation in Australia remains unclear. We looked at the effect of scenting traps with the faeces of the tiger quoll and the red fox on the capture rates of four species of CWR marsupials; the long- nosed bandicoot Perameles nasuta, the southern brown bandicoot Isoodon obesulus, the northern brown bandicoot I. macrourus and the common brushtail possum Chapter 3: Responses of CWR marsupials to predator odours 69

Trichosurus vulpecula. None of these species responded to fox odour. However, despite low capture rates, northern brown bandicoots and common brushtail possums were captured significantly more often in traps scented with tiger quoll faeces. This indicates that these species may be responding to predator odours, but not with the traditionally expected avoidance. Reasons for why this may be so are discussed.

3.2 Introduction

In the past 200 years, 16 species of mammals have become extinct in Australia, representing 50% of the total world mammalian extinctions over that time period.

(Short and Smith 1994). All of these extinctions have been species whose body mass lie within what has become termed the Critical Weight Range (CWR) of 35 g to 5500 g

(Burbidge and McKenzie 1989). Habitat loss, changes to fire regimes and competition have been proposed as possible causative agents. But predation by the red fox Vulpes vulpes which was successfully introduced to Australia in the 1870s (Rolls 1969), has been identified as a primary cause of the decline or extinction of 35 species which fall within the CWR (Burbidge and McKenzie 1989). The CWR coincides with the preferred prey-size range of the red fox (Henry 1996; Smith and Quin 1996), and the spread of the fox coincided with many recorded extinction events (Mahon 2001). It has been suggested that CWR species high daily metabolic requirements and/or low mobility may leave them more vulnerable to predation (Burbidge and McKenzie 1989;

Morton 1990). However, prior to the introduction of the fox, these CWR mammals were prey for native marsupial predators, such as the tiger quoll Dasyurus maculatus, to which they would have been just as vulnerable. So it is unclear why these CWR prey were seemingly so vulnerable to predation by introduced red foxes. Chapter 3: Responses of CWR marsupials to predator odours 70

Where predator and prey have co-evolved, natural selection has frequently led to an

‘arms race’ in detection and concealment strategies amongst enemies (Dawkins and

Krebs 1979). In particular, eavesdropping on predator signals has become a key weapon by which prey may estimate their predation risk. For example, many northern hemisphere mammalian species avoid the odour of their major mammalian predators

(Stoddart 1976; Dickman and Doncaster 1984; Gorman 1984; Calder and Gorman

1991). By avoiding areas contaminated with these odours they are thought to reduce the likelihood of encountering predators, and increase their chances of survival. In

Australia, native prey species have evolved with tiger quolls over millions of years, but foxes have only been present in Australia for the past 130 years (Rolls 1969; Mahon

2001). Indeed, Australia’s mammalian fauna has evolved in relative isolation for around 40 million years, and is dominated by marsupials, rather than placentals as throughout the rest of the world (Strahan 1998). Thus, potential prey may have defences which have evolved to reduce predation by marsupials such as tiger quolls, but these may not function for evolutionary novel predators like foxes (Russell et al.

2003). Similarly, defences which have evolved to reduce predation by foxes among northern hemisphere prey species may not have evolved in Australia at all. Moreover, predator odour recognition may not necessarily have evolved in the same way as in the rest of the world.

Evidence for the role of phylogeny or co-evolutionary history in the exploitation of odours in mammalian predator-prey interaction is currently lacking. It is not known whether marsupials avoid odours from either marsupial or placental predators and there is equivocal evidence in assessing whether placental prey species respond to marsupial predator odours or if co-evolution matters. Introduced house mice Mus domesticus Chapter 3: Responses of CWR marsupials to predator odours 71

avoid the faecal odour of the red fox, the feral cat Felis catus, and to a lesser degree the native western quoll D. geoffroii (Dickman 1992). However, Banks (1998) reported no aversion by two native species, the bush rat Rattus fuscipes and the agile antechinus

Antechinus agilis, to the faeces of the red fox. Similarly, Banks et al. (2003) found that domestic dog Canis lupus familiaris faeces at traps did not alter the capture rate of bush rats. Aversion to domestic dog urine has been reported for swamp wallabies Wallabia bicolor (Montague et al. 1990) and to synthetic predator odours for common brushtail possums Trichosurus vulpecula (Woolhouse and Morgan 1995; Morgan and

Woolhouse 1997), however in both cases this was determined by the consumption of seedlings that had been sprayed with the odour source, and this aversion may have been due to taste rather than odour. Blumstein et al (2002) found that the feeding choices of tammar wallabies Macropus eugenii and red-necked pademelons Thylogale thetis were not affected by red fox, brown bear Ursus arctos and dingo C. lupus dingo faeces or domestic dog urine, although Gresser (1996) did find that a combination of red fox urine and faeces decreased the amount of food eaten by common brushtail possums.

In this paper, we use a replicated field experiment to test whether four CWR marsupials, the long-nosed bandicoot Perameles nasuta, the southern brown bandicoot

Isoodon obeselus, the northern brown bandicoot I. macrourus and the common brushtail possum avoid odour cues from a native marsupial predator, the tiger quoll or an introduced placental predator, the red fox. Each species is preyed upon by both quolls and foxes (Alexander 1980; McKay 1994; Belcher 1995; Jones and Barmuta

1998), and all four species use scent-marking for intra-specific communication

(Stoddart 1980; Gordon 1998; Scott et al. 1999; Kerle 2001), indicating well developed olfactory abilities to support potential recognition of predators. We use the scat at trap Chapter 3: Responses of CWR marsupials to predator odours 72

technique (sensu Powell and Banks 2004) used in many northern hemisphere studies to demonstrate odour avoidance by potential prey. If these CWR mammals recognise and avoid predator odours as a predation risk, we predict that traps with faeces should have lower success than untreated traps.

3.3 Methods

Long-nosed bandicoots and common brushtail possums were trapped at the end of

August in Sydney Harbour National Park in Sydney, Latitude 33° 48’ S Longitude 151°

17’ E, at the boundary of thick heath and the open grassy areas around the old quarantine station located in the park. Tiger quolls are not found in the National Park

(Skelton et al. 2003; NPWS 2005), whilst red foxes enter every five years or so, killing large numbers of wildlife before being removed by park staff (Banks 2004). Cage traps

(50 x 19 x 19 cm), baited with bread and peanut butter and enclosed in green shade cloth to provide protection from the elements, were set in four transect lines, each consisting of six trap stations, for three consecutive nights on two separate occasions, with different transects being utilised on each occasion. Southern brown bandicoots and northern brown bandicoots were trapped during early September using large aluminium live-capture Elliott traps (46 x 16 x 16 cm), baited with a mixture of rolled oats, peanut butter and vegetable oil, and set out in a grid of 15 trap stations (3 x 5). Northern brown bandicoots were trapped in wet and dry heath in Myall Lakes National Park (latitude

32° 28’ S, longitude 152° 24’ E), three hours north of Sydney. Southern brown bandicoots were trapped in heath and open woodland in Garigal National Park (33° 43’

S, 151° 11’ E) in Sydney’s north. Both tiger quolls and red foxes are found in both

Myall Lakes and Garigal National Parks, although in Garigal tiger quolls appear to be extremely rare (McKay 1994; NPWS 1998; NPWS 2002; NPWS 2005). For these Chapter 3: Responses of CWR marsupials to predator odours 73

bandicoot species, trapping took place over three consecutive nights, and three grids were trapped on separate occasions. Whenever an animal was caught, the trap was removed and replaced with a fresh clean trap.

For all species, trap stations comprised of three traps placed one metre apart in a star formation, with stations 25-30 metres apart. At each station, one trap was scented with fox scat, one trap with quoll scat, and one trap was left untreated. This technique has been used extensively before to examine the issue of predator avoidance (Stoddart

1976; Calder and Gorman 1991; Dickman 1992; Banks 1998) and both Banks (1998) and our recent work on small mammals (Chapter 4) found no difference between the results of this technique and placing a single trap (either scented or unscented) at each trap station. Scented traps were treated by placing approximately 10 grams of scat across the entrance to the trap which emulated a typical encounter by these marsupials with such odours in the wild, and avoided any complications associated with trap dirtiness (Dickman 1992; Banks 1998). Both predator species use scats in scent marking (Macdonald 1979; Kruuk and Jarman 1995; Henry 1996; Burnett 2000); depositing scats in obvious and/or elevated positions along natural thoroughfares such as rocky creek beds or the bases of cliffs, as well as along man made thoroughfares such as roads (Belcher 1995; Kruuk and Jarman 1995; Triggs 1996; Burnett 2000).

Thus, scats are likely to be encountered by prey species, providing indications of the potential predator activity (Dickman 1992; Banks 1998).

Tiger quoll scats were collected from captive quolls held at Featherdale Wildlife Park in Sydney fed on a varied diet similar to what they would have in the wild, including poultry, rat, rabbit and macropods. Red fox scats were collected from captive foxes at Chapter 3: Responses of CWR marsupials to predator odours 74

the CSIRO Sustainable Ecosystems Unit in Canberra fed on a diet of sheep and kangaroo carcasses and dog food. Scats were either used fresh, or frozen for later use.

3.4 Results

Capture success was variable across the four species, though low as typical in studies of such medium-sized mammals. There were eight captures of southern brown bandicoots

(2.0%) and 11 captures of northern brown bandicoots from 405 trap-nights (2.7%), and

21 captures of long-nosed bandicoots (4.9%), and 13 captures of common brushtail possums from 432 trap-nights (3.0%). Most captures were of adult individuals (mean weight +/- SE); southern brown bandicoots, 699 r172 g; northern brown bandicoots

978 r 124 g; long-nosed bandicoots 800 r 202 g. There were also two recently independent young northern brown bandicoots weighing 160 g and 176 g, both trapped in quoll scented traps. In past studies, very few northern brown bandicoots of this size

(100-350 g) have been trapped (Hall 1983), and it has been assumed that animals of this size have a very high mortality (Gordon 1974; Hall 1983). Common brushtail possum weights were not measured to avoid unnecessary stress on animals; however all were full grown adults.

Chi-squared analysis showed no significant difference among treatments for captures of

2 2 long-nosed bandicoots (F 2=3.714, P=0.156) or southern brown bandicoots (F 2=3.250,

P=0.197). In contrast, there were differences in captures for northern brown bandicoots

2 2 (F 2=11.64, P=0.003) and common brushtail possums (F 2= 8.00, P=0.018) (Figure 1).

But contrary to predictions, captures for both species were higher in quoll scented traps than in untreated traps (Sign Test, P<0.05). No species showed any response to fox odours. The inclusion or exclusion of recaptures and captures when another trap at that Chapter 3: Responses of CWR marsupials to predator odours 75

trap station had been occupied had no effect on the lack of significant difference for long-nosed bandicoots and southern brown bandicoots as these captures were spread evenly among the three trap types, the same as the rest of the captures. One northern brown bandicoot capture was a recapture at a different trap station, but both captures of this animal were in quoll-scented traps. One northern brown bandicoot was captured in a quoll-scented trap at a trap station where an eastern chestnut mouse Pseudomys gracilicaudatus had been caught in a fox-scented trap on the same night. The unscented trap at this trap station remained open and unoccupied. Removal of these two captures

2 from the analysis still yielded a significant result (F 2= 8.00, P=0.018). Two captures of common brushtail possums in quoll-scented traps and the one capture in an unscented trap occurred when the fox-scented trap at that station was occupied by black rats

Rattus rattus for the quoll-scented traps and a long-nosed bandicoot for the unscented trap. The other trap at each of these stations remained open and unoccupied. One quoll- scented trap caught a common brushtail possum on two consecutive nights, and this may have been a recapture. The removal of these four captures from the analysis still

2 yielded a significant result (F 2= 6.00, P=0.050). Chapter 3: Responses of CWR marsupials to predator odours 76

unscented fox-scented 12 quoll-scented

10 * * 8

6

4

2

number of animals captured 0 P. nasuta I. obesulus I. macrourus T. vulpecula

Figure 3.1: The number of long-nosed bandicoots, southern brown bandicoots, northern brown bandicoots and common brushtail possums caught in unscented traps, traps scented with fox faeces, and traps scented with quoll faeces. * represents a significant difference from random entry of traps at the P=0.05 level.

3.5 Discussion

Neither the bandicoots nor possum showed any avoidance of fox faecal odours. This result contrasts with other studies of prey responses to fox odours where predator and prey have co-evolved (Dickman and Doncaster 1984; Calder and Gorman 1991).

However, it concurs with other Australian studies of rodents (Banks 1998) and macropods (Blumstein et al. 2002) which similarly showed no responses to fox faecal odours. Yet house mice in Australia, which have co-evolved with foxes, do avoid fox odours (Dickman 1992; but see Powell and Banks 2004). Together, these results support the notion that foxes are an evolutionarily novel predator, and their odours are not avoided by potential native prey in order to reduce their risks of predation (Banks

1998). Such naiveté may in part explain their particular vulnerability to foxes. Chapter 3: Responses of CWR marsupials to predator odours 77

However, these animals may use other behavioural means to reduce their likelihood of predation. For example, Gresser (1996) found that common brushtail possums ate fewer pieces of apple and sultana from areas tainted with red fox urine and faeces than untainted areas, so although they were not avoiding the area entirely, they were presumably spending less time there.

Similarly, none of these CWR marsupial species showed avoidance of native tiger quoll odours. Instead, northern brown bandicoots and common brushtail possums both appear to be attracted to the odour of the tiger quoll. Tiger quolls use scats in scent marking, often defecating in communal latrines (Kruuk and Jarman 1995). Latrine sites would represent particularly dangerous places, with frequent visitation by quolls. Kruuk and

Jarman (1995) recorded scat deposition rates of 1.35 to 1.9 scats per latrine per day, increasing to 3.4 scats per day when food was abundant. Daily visitation rates are likely even higher than this as tiger quolls do not always deposit scats when visiting latrines

(Claridge et al. 2004). The presence of quolls at the latrine several times per day would presumably lead to a higher risk of predation from the latrine user and other quolls attracted to the odours. Hence, it appears that these two species may be at higher risk of predation if they are attracted to quoll faecal odours.

Quoll scats may represent a possible source of invertebrates for bandicoots and possums. Dickman and Doncaster (1984) found that common shrews Sorex sorex did not avoid red fox scats, and suggested that scats may attract certain beetles, which the insectivorous shrews would prey upon to outweigh predation risks from foxes associated with fox scats. Bandicoots, while omnivorous, prefer to eat insects and other invertebrates (Quin 1988; Gott 1996; Scott et al. 1999) some of which may be Chapter 3: Responses of CWR marsupials to predator odours 78

associated with scats, and common brushtail possums are also known to consume invertebrates when readily available (Kerle 2001), but are otherwise largely herbivorous. Quoll scats may therefore be a highly profitable resource rich patch to be exploited by both species, whereas the similar value of fox scats may not be recognised because they are recent arrivals to Australia. However, it is not clear why this would be the case for the northern brown bandicoot and the far less insectivorous common brushtail possum, and not for the more insectivorous long-nosed and southern brown bandicoots.

It is also possible that bandicoots and possums showed an anti-predator response in spite of the apparent attraction to traps with odours. In response to a perceived predation threat that may be imminent, prey can either opt for flight or concealment.

Some rodents exposed to predators take flight and flee while others may freeze

(Jedrzejewski et al. 1992). But both in the wild and in the lab, imminent predator threat causes many animals to take refuge in the nearest refugium, typically somewhere where predators cannot follow (Jedrzejewski et al. 1992; Dielenberg et al. 1999). It may be that these species respond to predator odours, by entering the nearest appropriately sized hole, which in this case is a trap. Both northern brown bandicoots and common brushtail possums nest in natural hollows (Gordon 1998; Kerle 2001) and take shelter when approached.

But why do long-nosed and southern brown bandicoots not respond to quoll odour in the same way as northern brown bandicoots and common brushtail possums? One possible reason is size-dependent risks of predation from quolls influencing prey responses to odours. Common brushtail possums may weigh more than 4 kg, and Chapter 3: Responses of CWR marsupials to predator odours 79

northern brown bandicoots may weigh more than 3 kg, whereas both the southern brown bandicoot and long-nosed bandicoot seldom weigh more than 1 kg (Strahan

1998). As such, these smaller species are likely to have been vulnerable to predation by eastern quolls D. viverrinus before this native predator became extinct on the mainland, as eastern quolls are thought to prey on animals up to a size of 1.5 kg (Jones and Rose

2001), whereas the larger prey would only be at risk from tiger quolls. Small prey would therefore be expected to strongly avoid areas with odours from any quoll species while larger prey should show more species specific responses to predator odours

(Jedrzejewski et al. 1993) to avoid costs of unnecessary odour avoidance. In support of this concept we found that even allowing for a single trap at each trap station, trap success for long-nosed bandicoots was lower than the long term average success for the population (14.6% vs 22%) (NPWS, unpublished data) suggesting some avoidance of the trap stations as opposed to finer scale odour discrimination. In contrast, the apparent attraction of larger prey species to quoll odours may represent a strategy of odour inspection to discriminate amongst odour donors, information which is often found in the less volatile components (Nevison et al. 2003) requiring closer contact. But once the predation risk was identified, the larger prey took refuge. To test this hypothesis further, finer scale information is needed about the immediate behavioural responses to predator odours by these prey species.

None of these four CWR species avoided the odour of the introduced red fox. Such apparent naiveté to fox odour cues may put them at higher risks of predation than introduced species which do avoid fox odours, and may in part explain their particular vulnerability to fox predation. However, they may respond in other ways which do not affect their trappability. Common brushtail possums and northern brown bandicoots Chapter 3: Responses of CWR marsupials to predator odours 80

appear sensitive to odours from native tiger quolls, but may be adopting a strategy of predator inspection. Future work on these odour mediated predator prey interactions will benefit from resolving the fine scale behavioural responses of prey to different odour sources.

3.6 Acknowledgements

We would like to thank Alan Russell, Tim Chapman and Jeremy Green for their help in the field. Thanks to Rachel Miller and National Parks Staff for their help at North Head.

Thank you to Rob Humphries (DEC) for access to data on long-nosed bandicoots at

North Head. Thanks also to Steve Henry and the CSIRO vertebrate pests unit for supplying fox faeces, and to Brad Walker, Chad Staples and Featherdale Wildlife Park for supplying quoll faeces. We thank Bruce Mitchell and Barry Fox for their constructive comments on an earlier draft of this manuscript. Chapter 3: Responses of CWR marsupials to predator odours 81

3.7 References

Alexander, D. (1980). Diet of the Tiger Quoll (Dasyurus maculatus Kerr) in an area of New England National Park, Northeastern New South Wales. Honours thesis, University of New England, Armidale.

Banks, P. B. (1998). Responses of Australian Bush Rats, Rattus fuscipes, to the odour of introduced Vulpes vulpes. Journal of Mammalogy 79, 1260-1264.

Banks, P. B. (2004). Population viability analysis in urban wildlife management: modelling management options for Sydney's quarantined bandicoots. In Urban Wildlife: more than meets the eye. (edited by D. Lunney and S. Burgin) pp. 70- 77. Royal Zoological Society of New South Wales, Mosman.

Banks, P. B., Hughes, N. K. and Rose, T. A. (2003). Do native Australian small mammals avoid faeces of domestic dogs? Responses of Rattus fuscipes and Antechinus stuartii. Australian Zoologist 32, 406-409.

Belcher, C. A. (1995). Diet of the tiger quoll (Dasyurus maculatus) in East Gippsland, Victoria. Wildlife Research 22, 341-357.

Blumstein, D. T., Mari, M., Daniel, J. C., Ardon, J. G., Griffith, A. S. and Evans, C. S. (2002). Olfactory predator recognition: wallabies may have to learn to be wary. Animal Conservation 50, 143-198.

Burbidge, A. A. and McKenzie, N. L. (1989). Patterns in the modern decline of Western Australia's vertebrate fauna: causes and conservation implications. Biological Conservation 50, 143-198.

Burnett, S. E. (2000). Ecology and Conservation Status of the northern Spot-tailed Quoll, Dasyurus maculatus: with reference to the Future of Australia's Marsupial Carnivores. Ph.D. dissertation, James Cook University of Northern Queensland, Townsville: 292.

Calder, C. J. and Gorman, M. L. (1991). The effects of Red Fox Vulpes vulpes faecal odours on the feeding behaviour of Orkney Voles Microtus arvalis. Journal of Zoology (London) 224, 599-606.

Claridge, A. W., Mifsud, G., Dawson, J. and Saxon, M. J. (2004). Use of infrared digital cameras to investigate the behaviour of cryptic species. Wildlife Research 31, 645-650.

Dawkins, R. and Krebs, J. R. (1979). Arms races between and within species. Proceedings of the Royal Society of London B 205, 489-511.

Dickman, C. R. (1992). Predation and Habitat Shift in the House Mouse, Mus domesticus. Ecology 73, 313-322.

Dickman, C. R. and Doncaster, C. P. (1984). Responses of Small Mammals to Red Fox (Vulpes vulpes) Odour. Journal of Zoology (London) 204, 521-531. Chapter 3: Responses of CWR marsupials to predator odours 82

Dielenberg, R. A., Arnold, J. C. and McGregor, I. S. (1999). Low-Dose Midazolom Attenuates Predatory Odor Avoidance in Rats. Pharmacology Biochemistry and Behaviour 62, 197-201.

Gordon, G. (1974). Movements and Activity of the Shortnosed Bandicoot Isoodon macrourus Gould (Marsupialia). Mammalia 38, 405-431.

Gordon, G. (1998). Northern Brown Bandicoot Isoodon macrourus. In The Mammals of Australia. (edited by R. Strahan) pp. 174-175. New Holland Publishers, Sydney.

Gorman, M. L. (1984). The response of prey to Stoat (Mustela erminea) scent. Journal of Zoology (London) 202, 419-423.

Gott, M. (1996). Ecology of the Northern Brown Bandicoot, Isoodon macrourus: reproduction and resource use in a Heathland population. Ph.D. dissertation, University of New South Wales, Sydney.

Gresser, S. (1996). Anti-predator Behaviour of the Common Brushtail Possum (Trichosurus vulpecula) at Burrendong Dam, N.S.W. Honours Thesis, University of Sydney, Sydney: 106.

Hall, L. S. (1983). Observations on Body Weights and Breeding of the Northern Brown Bandicoot, Isoodon macrourus, Trapped in South-East Queensland. Australian Wildlife Research 10, 467-476.

Henry, J. D. (1996). Red Fox: The Catlike Canine. Smithsonian Institute Press, Washington D.C.

Jedrzejewski, W., Jedrzejewska, B. and McNeish, E. (1992). Hunting success of the weasel Mustela nivalis and escape tactics of forest rodents in Bialowieza National Park. Acta Theriologica 37, 319-328.

Jedrzejewski, W., Rychlik, W. and Jedrzejewska, B. (1993). Responses of bank voles to odours of seven species of predators: experimental data and their relevance to natural predator-vole relationships. Oikos 68, 251-257.

Jones, M. E. and Barmuta, L. A. (1998). Diet overlap and abundance of sympatric dasyurid carnivores: a hypothesis of competition? Journal of Animal Ecology 81, 434-447.

Jones, M. E. and Rose, R. K. (2001). Dasyurus viverrinus. Mammalian Species 677, 1- 9.

Kerle, J. A. (2001). Possums: the brushtails, ringtails and greater glider. University of New South Wales Press, Sydney.

Kruuk, H. and Jarman, P. J. (1995). Latrine use by the Spotted-tailed quoll (Dasyurus maculatus: Dasyuridae, Marsupialia) in its natural habitat. Journal of Zoology (London) 236, 345-349. Chapter 3: Responses of CWR marsupials to predator odours 83

Macdonald, D. W. (1979). Some Observations and Field Experiments on the Urine Marking Behaviour of the Red Fox Vulpes vulpes L. Zeitshrift fuer Tierpsychologie 51, 1-22.

Mahon, P. (2001). Draft Threat Abatement Plan for Predation by the Red Fox (Vulpes vulpes). NSW National Parks and Wildlife Service, Hurstville.

McKay, G. M. (1994). Effects of introduced predators on biodiversity remnant vegetation. Australian Nature Conservation Agency, Canberra.

Montague, T. L., Pollock, D. C. and Wright, W. (1990). An examination of the Browsing animal problem in Australian eucalypt and pine plantations. Proceedings of Vertebrate Pest Conferences 14, 203-208.

Morgan, D. R. and Woolhouse, A. D. (1997). Predator Odours as repellents to Brushtail Possums and Rabbits. In Repellents in Wildlife Management. (edited by J. R. Mason) pp. 241-252. National Wildlife Research Center, Ft. Collins.

Morton, S. R. (1990). The impact of European Settlement on the vertebrate animals of arid Australia: a conceptual model. Proceedings of the Ecological Society of Australia 16, 210-213.

Nevison, C. M., Armstrong, S., Beynon, R. J., Humphries, R. E. and Hurst, J. L. (2003). The ownership signature in mouse scent marks is involatile. Proceedings of the Royal Society of London B 270, 1957-1963.

NPWS (1998). Garigal National Park Plan of Management. Hurstville, NSW National Parks and Wildlife Service: 46.

NPWS (2002). Myall Lakes National Park, Little Broughton Island and Stormpetrel Nature Reserves Plan of Management. Hurstville, NSW National Parks and Wildlife Service: 148.

NPWS (2005). Atlas of NSW Wildlife, NSW National Parks and Wildlife Service. 2005.

Powell, F. and Banks, P. B. (2004). Do house mice modify their foraging behaviour in response to predator odours and habitat? Animal Behaviour 67, 753-759.

Quin, D. G. (1988). Observations on the diet of the Southern Brown Bandicoot, Isoodon obesulus (Marsupialia: Peramelidae) in Southern Tasmania. Australian Mammalogy 11, 15-25.

Rolls, E. C. (1969). They All Ran Wild, the Animals and Plants that Plague Australia. Angus and Robertson, Sydney.

Russell, B. G., Smith, B. and Augee, M. L. (2003). Changes to a population of common ringtail possums (Pseudocheirus peregrinus) after bushfire. Wildlife Research 30, 389-396. Chapter 3: Responses of CWR marsupials to predator odours 84

Scott, L. K., Hume, I. D. and Dickman, C. R. (1999). Ecology and population biology of Long-nosed Bandicoots (Perameles nasuta) at North Head, Sydney Harbour National Park. Wildlife Research 26, 805-821.

Short, J. and Smith, A. (1994). Mammal decline and recovery in Australia. Journal of Mammalogy 75, 288-297.

Skelton, N., Richmond, O., Gilson, T. and Wong, P. (2003). Fauna of North Head; including the lands at the Former School of Artillery, the Royal Australian Artillery National Museum, Sydney Harbour National Park and the Australian Institute of Police Management. Sydney Harbour Federation Trust, Sydney.

Smith, A. P. and Quin, D. G. (1996). Patterns and causes of extinction and decline in Australian conilurine rodents. Biological Conservation 77, 243-267.

Stoddart, D. M. (1976). Effect of Odour of Weasels (Mustela nivalis L.) on Trapped Samples of Their Prey. Oecologia 22, 439-441.

Stoddart, D. M. (1980). Observations of the Structure and Function of Cephalic Skin Glands in Bandicoots (Marsupialia: Peramelidae). Australian Journal of Zoology 28, 33-41.

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

Triggs, B. (1996). Tracks, scats and other traces; A field guide to Australian mammals. Oxford University Press, South Melbourne.

Woolhouse, A. D. and Morgan, D. R. (1995). An evaluation of repellents to suppress browsing by possums. Journal of Chemical Ecology 21, 1571-1583. Chapter 4: Do Australian small mammals avoid predator odours? 85

Chapter 4:

Do Australian small mammals avoid native and introduced predator odours?

4.1 Abstract

Co-evolution is thought to have led to many small mammal species avoiding the scent marks of their mammalian predators as they provide a reliable cue to predation risk.

Most support for this hypothesis comes from northern hemisphere predator/prey systems, however it is unclear whether this avoidance of predator odour occurs in

Australia’s mammalian fauna, which has evolved in relative isolation from the rest of the world, and is dominated by marsupials rather than placentals. I tested this theory for an Australian system with marsupial and placental predators and prey, that share a long term (>1million years) or short term exposure (<150 years) to each other. The predators were the native marsupial tiger quoll Dasyurus maculatus and the introduced placental red fox Vulpes vulpes. The potential prey were three native rodent species, the bush rat

Rattus fuscipes, the swamp rat R. lutreolus, the eastern chestnut mouse Pseudomys gracilicaudatus, and the marsupial brown antechinus Antechinus stuartii. Small mammals were captured in Elliott traps with 1/3 of traps treated with fox faeces, 1/3 treated with quoll faeces and the remainder left untreated. The native rodent species all showed avoidance of both tiger quoll and red fox odours whereas the marsupial brown antechinus showed no responses to either odour. Either predator odour avoidance has not evolved in this marsupial or their reaction to predator odours may be exhibited in ways which are not recognisable through trapping. The avoidance by the rodents of fox Chapter 4: Do Australian small mammals avoid predator odours? 86

odour as well as quoll odour suggest this response may either be due to common components in fox and quoll odour, or it may be a rapidly evolved response.

4.2 Introduction

Scent marking with a combination of urine, faeces or glandular secretions is a common behaviour which has been observed in all carnivore families (Gorman and Trowbridge

1989). Scent marks are typically deposited on conspicuous objects along well-traversed pathways, and frequently remarked (Kleiman 1966; Macdonald 1980). Their odour may convey information about identity, territorial boundaries, reproductive state, and social status (Gorman and Trowbridge 1989; Rostain et al. 2004). But this information can also be exploited by “eavesdropping” prey species. In the northern hemisphere, many species of small mammal consistently avoid traps tainted with the scent-marks of the mammalian predators with which they have co-evolved (Stoddart 1976; Dickman and

Doncaster 1984; Gorman 1984). Visitation by con-specifics and regular remarking means that scent marked areas may become focal points for predator activity (Kleiman

1966; Gorman 1980; Macdonald 1980; Gorman and Trowbridge 1989). So by avoiding these areas prey species reduce their likelihood of predator encounter. This odour- sensitive avoidance has the potential to deter potential prey from feeding patches, leading to potential indirect or sub-lethal effects of predation on prey populations, but these prey do avoid the ultimate fitness cost of death (Lima 1998).

In Australia, odour-induced predator avoidance has received only limited attention, mostly associated with concerns over the impacts of introduced predators and prey. For example, in Western Australia introduced house mice Mus domesticus avoid traps scented with the faeces of the red fox Vulpes vulpes and the feral cat Felis catus with Chapter 4: Do Australian small mammals avoid predator odours? 87

which they have co-evolved, but they also avoid traps scented with the faeces of the evolutionarily unfamiliar native western quoll Dasyurus geoffroii, albeit to a lesser extent (Dickman 1992). In contrast, Banks (1998) reported no aversion by native bush rats Rattus fuscipes and agile antechinus Antechinus agilis to entering traps scented with red fox faeces and Banks et al. (2003) found no reduction in the capture rate of bush rats due to the faeces of the domestic dog Canis lupus familiaris. Although these native prey species did not avoid the faecal odour of these evolutionarily unfamiliar predators, it is unknown how they respond to the scent marks of native marsupial predators with which they have coevolved.

In Chapter 3, I presented the first study of Australian prey species responses to native marsupial predators. I found that although common brushtail possums Trichosurus vulpecula and northern brown bandicoots Isoodon macrourus did not respond to the faecal odour of the introduced red fox, they did respond to the faecal odour of the native tiger quoll D. maculatus, but with apparent attraction rather than avoidance.

While this behaviour may have adaptive value in reducing predation risk from quolls

(Russell and Banks 2005), this attraction is at odds with earlier observed general responses of mammal species, mostly rodents. There are a number of possible explanations for this difference. Firstly, northern brown bandicoots and common brushtail possums are almost two orders of magnitude larger than the rodents which were trapped in earlier odour-avoidance studies (Table 4.1). Avoidance strategies which are appropriate for smaller species may not be as appropriate for larger prey species.

Secondly, these marsupial prey have coevolved with marsupial predators over the 40 million years of Australia’s isolation from the rest of the world (Strahan 1998). It is possible that they developed different responses to predator odours compared to their Chapter 4: Do Australian small mammals avoid predator odours? 88

placental counterparts evolving throughout the rest of the world. Finally, the differences in hunting and scent marking strategies between quolls and placental predators may induce different responses to their odour, in which case both smaller marsupials and rodents may also be caught more frequently in quoll-scented traps.

Table 4.1: Comparison of the weights of northern brown bandicoots and common brushtail possums versus the weights of rodent species which have exhibited predator odour avoidance in trapping studies. Species Weight Study Woodmouse Apodemus 13-30 g (Dickman and Doncaster sylvaticus * 1984)

Bank vole Clethrionomys 14-40 g (Dickman and Doncaster glareolus * 1984)

Field Vole Microtus 14-50 g (Stoddart 1976; Gorman 1984) agrestis *

Gray-tailed Vole M. 35-55 g (Wolff and Davis-Born 1997) canicaudus **

Orkney Vole M. arvalis 14-100 g (Gorman 1984; Calder and * Gorman 1991)

Northern Brown Bandicoot 500-3100 g Chapter 3 *** Common Brushtail Possum 1200-4500 g Chapter 3 *** * weights taken from Macdonald and Barrett (1993) ** weights taken from Kays and Wilson (2002) *** weights taken from Strahan (1998)

This chapter examines the responses of native small mammal species to the faecal odours of native tiger quolls and introduced red foxes by live-trapping. Both tiger quoll and red fox faeces were used so that direct comparisons could be made between the responses to the two predator odours. Responses of three rodent species; the bush rat, the swamp rat R. lutreolus, and the eastern chestnut mouse Pseudomys gracilicaudatus, Chapter 4: Do Australian small mammals avoid predator odours? 89

and one marsupial dasyurid; the brown antechinus A. stuartii were measured. Remains of all four species are consistently found in the scats of both predator species

(Alexander 1980; Green and Osbourne 1981; Lunney et al. 1990; McKay 1994;

Belcher 1995; Triggs 1996; Mitchell and Banks 2005). Tiger quoll and red fox faeces were the scent marks used in this experiment as they are likely to be encountered by prey species in the field, being deposited on conspicuous places throughout the territories of these animals such as elevated areas and intersections and corners of tracks and roads (Belcher 1995; Kruuk and Jarman 1995; Triggs 1996; Burnett 2000), and therefore potentially provide information about the predator’s identity and whereabouts (Dickman 1992; Banks 1998).

4.3 Study area

The main study site was in Myall Lakes National Park (latitude 32º 28’ S, longitude

152º 24’ E) approximately 250km north of Sydney, NSW, Australia. Both tiger quolls and red foxes inhabit the area and prey upon a variety of native species (McKay 1994;

NPWS 2002; NPWS 2005). Two basic habitat types were sampled; open eucalypt forest and heath. The open eucalypt forest was dominated by Angophora costata and

Eucalyptus pilularis. The heath areas were mainly wet heath grading into dry heath and swamp, dominated by Banksia oblongifolia, Dillwynia floribunda, Leptospermum liversidgei and L. juniperum. Bush rats were mainly captured in the open eucalypt forest, whereas swamp rats and eastern chestnut mice were mainly captured in the heath areas, and brown antechinus were captured equally in both habitats. Additional study sites were also trapped closer to Sydney; in Garigal National Park (latitude 33º 45’ S, longitude 151º 12’ E) 15km north of Sydney’s central business district (CBD), and

Royal National Park (latitude 34º 10’ S, longitude 151º 05’ E) 30km south of the CBD. Chapter 4: Do Australian small mammals avoid predator odours? 90

Whereas the red fox is found throughout both national parks, the tiger quoll is extremely rare in Garigal National Park and is absent in Royal NP (NPWS 1998;

NPWS 2000; NPWS 2005). The habitats trapped in these areas were eucalypt woodland dominated by A. costata and E. haemastoma, and riparian temperate rainforest, dominated by Ceratopetalum apetalum. Only bush rats and brown antechinus were caught in these Sydney sites.

4.4 Methods

The responses of small mammals to predator faecal odours were examined through live-trapping. Collapsible aluminium Elliot traps were baited with a mixture of rolled oats, peanut butter and vegetable oil. Treatment traps were then scented by leaving an approximately equal volume of faeces across the entrance to each trap. This approach was used in favour of lining the inside of the trap with faeces in order to avoid the need to control for trap dirtiness (Dickman 1992; Banks 1998). Tiger quoll faeces were collected from captive animals at Featherdale Wildlife Park in Sydney fed a varied diet similar as to what they would have in the wild, including poultry, rat, rabbit, and macropod. Red fox faeces were collected from captive animals at the CSIRO

Sustainable Ecosystems Unit in Canberra that were fed on a diet of sheep and kangaroo carcasses and dog food. Faeces were either used fresh, or frozen for later use. Many studies have attempted a procedural control to control for the addition of “strong” odour, but there is no odour that is strong without having unique scents associated with it, and hence traps without faeces were used as “controls” as they carried the same types of residual odours as the treated traps excluding the predator treatment (Banks 1998). Chapter 4: Do Australian small mammals avoid predator odours? 91

4.4.1 Sampling designs

Two sampling designs were used to account for possible artefacts in the way treatments were presented to prey. In the first design, sampling sites consisted of 15 trap stations each with a single small Elliott Type B trap (33 x 10 x 9 cm) and spaced 10-15 m apart.

One third of these traps were scented with quoll faeces, one third were scented with fox faeces and one third were left unscented. The sequence of fox-scented, quoll-scented, and unscented traps remained constant at each site so that traps with the same odour were always equally spaced apart, but the sequence was changed at each site. In the second trap design, three large Elliott traps (46 x 16 x 16 cm) 75 cm apart were placed at each of 15 trap stations 25-30 m apart, with one trap being scented with quoll faeces, one trap being scented with fox faeces and one trap left unscented. The results of this multiple trap design may be considered to express the responses of small mammals when given the choice between scented and unscented traps and has been used extensively to test mammal responses to predator odours (Stoddart 1976; Dickman and

Doncaster 1984; Calder and Gorman 1991; Dickman 1992; Banks 1998). However, it is possible prey individuals might avoid the trap station altogether, due to the relatively close proximity of the predator-scented traps and the unscented traps. There is also a possible issue of non-independence of treatment and control if mild avoidance of predator odour induces animals to immediately enter an available adjacent clean trap thereby falsely exaggerating predator avoidance (Banks 1998). The results of the single trap design, on the other hand, express the response of the small mammals upon encountering these odours in the field, and might be a more reliable indicator of the likelihood of an animal to give up foraging in order to avoid the odour of these predators. I therefore evaluated both sampling designs to compare the results between the two for any differences in the responses of the small mammal species. Chapter 4: Do Australian small mammals avoid predator odours? 92

In total, 36 sites were trapped using the single trap design and 14 sites were trapped using the three trap design. However, only sites where individuals of a species were captured were included in the analysis for that species. For both sampling designs, half the sites were trapped during Spring (late August to early November) and half were trapped during Summer (early December to early March). Sites were separated by a distance of at least 300 m to facilitate independence of replicates. For both designs, each site was trapped for three nights, and traps were left open throughout the entire trapping period. Animals captured were sexed, weighed, measured, and marked on the ear with a water resistant ink marker to identify recaptures. Animals were classified as either adult or juvenile. Adults were defined on the basis of weight ranges given by

Strahan (1998). Used traps were replaced with clean traps, which had been thoroughly scrubbed with mild detergent and rinsed.

4.4.2 Data analysis

Only the first captures of individuals were included in the analyses, to avoid issues of pseudoreplication of the same individual, i.e. recaptures were not included in the analysis. Data was used from all three nights for each trap line, as new individuals were caught on each day at each sampling site. The numbers of captures of each species were analysed using two methods which addressed slightly different issues regarding prey responses to the treatments. The first was using an unreplicated Analysis of Variance

(ANOVA) to compare the number of captures in unscented, quoll-scented and fox- scented traps between the different sampling sites. In this method, the two designs were analysed separately, as there were 45 traps per site in the three trap design, as opposed to 15 traps per site in the single trap design, but accounted for any site differences. Chapter 4: Do Australian small mammals avoid predator odours? 93

Initially, season was included as a factor in this analysis, but the odour by season interaction was non-significant for all four species, and season was removed from the analysis. The second method was using Binary Logistic Regression, which allowed for direct comparison between the sampling designs, and for bush rats and brown antechinus, between the Myall Lakes sites where tiger quolls are still common and the

Sydney sites where tiger quolls are rare or absent. In this method, replicates were the individual traps. Traps were pooled across sites, as site, being a random factor, could not be included as a predictor in the model. The response was either capture or no capture, and the predictors were odour (unscented, quoll-scented, or fox-scented), design (single trap or three trap), and for bush rats and brown antechinus, location

(Myall Lakes or Sydney). This method also took trap availability into account, as traps which were closed, disturbed, or caught recaptures or other species were removed from the analysis. As Dickman and Doncaster (1984) found that the weights of some rodent species differed between animals caught in fox-scented and unscented traps, the distribution of the weights of each species were analysed by two factor ANOVA, with odour (unscented, quoll-scented, or fox-scented) and design (single trap or three trap) as factors.

4.5 Results

In total, 200 bush rats were captured from 28 sites; 64 swamp rats were captured from

24 sites; 59 eastern chestnut mice were captured from 17 sites; and 73 brown antechinus were captured from 25 sites. Chapter 4: Do Australian small mammals avoid predator odours? 94

4.5.1 Bush Rats

The presence of the predator odours had a significant effect on the number of captures of bush rats in both the three trap design and the single trap design (Table 4.2a). In both designs significantly fewer bush rats were caught in fox-scented traps and quoll-scented traps than in unscented traps, and there was no difference between the number of captures in fox-scented and quoll-scented traps (Figure 4.1). There were no differences between the numbers of captures in the Myall Lakes sites compared to the Sydney sites, no interaction between odour and location, and although there was a significant difference between the capture rates of the two designs, there was no odour by design interaction (Table 4.3a).

4.5.2 Swamp Rats

As for bush rats, the presence of the predator odours had a significant effect on the number of captures of swamp rats in the single trap design, and a strong trend towards an effect in three trap design (Table 4.2b). But unlike for bush rats, there was an odour by design interaction (Table 4.3b). In the three trap design, there were equal captures of swamp rats in unscented and fox-scented traps, but with a strong trend towards fewer captures quoll-scented traps (p=0.061) (Figure 4.1a). However, in the single trap design, as for bush rats, fewer swamp rats were caught in fox-scented traps and quoll- scented traps compared to unscented traps, and there was no difference between the number of captures in fox-scented and quoll-scented traps (Figure 4.1b).

4.5.3 Eastern Chestnut Mice

Similarly to bush rats, in the three trap design significantly fewer eastern chestnut mice were trapped in quoll-scented and fox-scented traps than in unscented traps (Figure Chapter 4: Do Australian small mammals avoid predator odours? 95

4.1a), but unlike bush rats or swamp rats the single trap design revealed no effect of predator odours on captures (Figure 4.2b, Table 4.2c). However, the binary logistic regression showed that there was no difference in the number of captures between the two designs and no odour by design interaction, and an overall strong trend towards avoidance of both predator odours (Table 4.3c).

4.5.4 Brown Antechinus

In contrast to the three rodent species, neither of the predator odours affected captures of brown antechinus in either trap design (Table 4.2d; Figure 4.1). Similarly there was no effect of design or location on the number of antechinus captured (Table 4.3d)

4.5.5 Weights

The average weight of individuals of each species captured in unscented, fox-scented and quoll-scented traps was not significantly different (Fig 4.2; Table 4.4).

Table 4.2: Results of unreplicated ANOVAs comparing the number of captures per site in unscented, fox-scented and quoll-scented traps of four species of small mammal, using the single trap and three trap sampling designs. Odour Site

(a) Bush Rats

Three trap design F2,14=5.55 P=0.017 F7,14=3.92 P=0.014

Single trap design F2,38=6.03 P=0.005 F19,38=2.40 P=0.011

(b) Swamp Rats

Three trap design F2,12=3.55 P=0.062 F6,12=3.25 P=0.039

Single trap design F2,32=6.57 P=0.004 F16,32=1.41 P=0.197 Chapter 4: Do Australian small mammals avoid predator odours? 96

(c) Eastern Chestnut Mouse

Three trap design F2,10=6.06 P=0.019 F5,10=1.65 P=0.233

Single trap design F2,20=2.03 P=0.158 F10,20=1.07 P=0.429

(d) Brown Antechinus

Three trap design F2,14=0.06 P=0.928 F7,14=0.68 P=0.689

Single trap design F2,32=0.40 P=0.675 F16,32=0.45 P=0.955 Chapter 4: Do Australian small mammals avoid predator odours? 97

(a) three trap design

7 a unscented 6 fox-scented 5 d quoll-scented 4 bb 3 c e 2 c ff f c* e captures per site 1 0 Bush Rat Swamp Rat Eastern Brown Chestnut Antechinus Mouse

(b) single trap design

3.5 a unscented 3 fox-scented quoll-scented 2.5 b b e 2 c 1.5 f d e e f 1 f d

captures per site captures 0.5 0 Bush Rat Swamp Rat Eastern Brown Chestnut Antechinus Mouse

Fig 4.1: The mean number (± standard error) of bush rats, swamp rats, eastern chestnut mice, and brown antechinus caught per site in unscented, fox-scented and quoll-scented traps using (a) the three trap design, and (b) the single trap design. Different letters above the bars for each species indicate a significant difference (P<0.05). c* indicates that although not significant there was a strong trend towards a difference (P=0.061). Chapter 4: Do Australian small mammals avoid predator odours? 98

Table 4.3: Results of Binary Logistic Regressions comparing the number of captures in unscented, fox-scented and quoll-scented traps between the three trap and single trap sampling designs, and (for bush rats and brown antechinus) between locations (Myall Lakes or Sydney) for four species of Australian small mammal. Test Statistic df P

(a) Bush Rats Odour Ȥ2=12.84 2 0.002 Location z=0.68 1 0.493 Design z=3.02 1 0.003 Location*Design z=1.57 1 0.116 Odour*Location Ȥ2=2.09 2 0.351 Odour*Design Ȥ2=0.87 2 0.657 Odour*Location*Design Ȥ2=1.03 2 0.742

(b) Swamp Rats Odour Ȥ2=13.31 2 0.001 Design z=3.43 1 0.001 Odour*Design Ȥ2=6.27 2 0.044

Eastern Chestnut Mice Odour Ȥ2=4.73 2 0.094 Design z=0.85 1 0.393 Odour*Design Ȥ2=3.41 2 0.182

Brown Antechinus Odour Ȥ2=0.06 2 0.969 Location z=1.37 1 0.209 Design z=1.79 1 0.074 Location*Design z=0.20 1 0.843 Odour*Location Ȥ2=3.20 2 0.202 Odour*Design Ȥ2=0.20 2 0.906 Odour*Location*Design Ȥ2=2.76 2 0.252 Chapter 4: Do Australian small mammals avoid predator odours? 99

Table 4.4: Results of two-way ANOVAs comparing the weights of animals caught in unscented, fox-scented and quoll-scented traps for bush rats, swamp rats, eastern chestnut mice and brown antechinus. Odour Design Odour*Design

(a) Bush Rat

F2,203=1.16 P=0.315 F1,203=11.01 P=0.001 F2,203=2.18 P=0.315

(b) Swamp Rat

F2,51=1.03 P=0.363 F1,51=0.29 P=0.595 F2,51=1.01 P=0.371

(c) Eastern Chestnut Mouse

F2,45=1.34 P=0.271 F1,45=3.61 P=0.064 F2,45=0.76 P=0.475

( d) Brown Antechinus

F2,56=0.82 P=0.447 F1,56=1.45 P=0.234 F2,56=0.59 P=0.559

160 unscented 140 fox-scented quoll-scented 120 100 80 60

weight (grams) weight 40 20 0 Bush Rat Swamp Rat Eastern Brown Chestnut Antechinus Mouse

Fig 4.2: The mean weight in grams (± standard error) of bush rats, swamp rats, eastern chestnut mice, and brown antechinus caught in unscented, fox-scented and quoll- scented traps. Chapter 4: Do Australian small mammals avoid predator odours? 100

4.6 Discussion

All three native rodent species were caught in traps scented with the faecal odour of the tiger quoll less frequently than they were in unscented traps. These results indicate avoidance of native predator odours has evolved in at least some Australian small mammal species, and that this may be an effective method for reducing their risk of predation from these native Australian predators. Avoidance did not appear to be reliant upon current predation pressure by tiger quolls in the study area, as there was no odour by location interaction. Tiger quoll odour was avoided in both the Myall Lakes sites where tiger quolls are relatively common (NPWS 2002; NPWS 2005), and the Sydney sites where they are extremely rare or absent (NPWS 1998; NPWS 2000; NPWS 2005).

Red fox faecal odour also deterred these rodents, and avoidance was strong and similar for the two predator species. Captures of bush rats in quoll-scented and fox-scented traps were 53% and 52% of those in unscented traps, captures of swamp rats in quoll- scented and fox-scented traps were 43% and 40% of those in unscented traps, and captures of eastern chestnut mice in quoll-scented and fox-scented traps were 32% and

38% of those in unscented traps. The strength and similarity of the rodents’ responses to tiger quoll odour and red fox odour is somewhat surprising given that rodents share a long co-evolution with tiger quolls, but have been in contact with red foxes for less than

130 years (Rolls 1969). The common response may be due to volatile components common to the odours of these two predators (Dickman and Doncaster 1984; Dickman

1992). There is a much higher degree of evolutionary separation between tiger quolls and red foxes than between other carnivore species which have been found to share sulfurous odour components aversive to prey species. These species have all been closely related members of the placental order Carnivora (see Chapter 1). However, it Chapter 4: Do Australian small mammals avoid predator odours? 101

has been suggested that these aversive odour components may be commonly derived from a carnivorous diet (Nolte et al. 1994).

In the only other study to use marsupial predator odours, Dickman (1992) found that in

Western Australia introduced house mice avoided traps scented with the faeces of the evolutionarily unfamiliar western quoll. However, the level of avoidance was weaker than for traps scented with the faeces of the red fox and the feral cat, predators with which house mice have co-evolved (Dickman 1992). This result suggests that if there are common odour components which trigger a generalised carnivore response

(Dickman and Doncaster 1984), prey species may adjust this generalised response according to other components found in the odours of each predator species. This is in accordance with the results of Jedrzejewski et al. (1993), who found that although bank voles Clethrionomys glareolus responded to the odours of six different mammalian predators, the response to each predator was subtly different.

It was suggested that the weaker avoidance of western quoll faecal odour by house mice, compared to their avoidance of red fox and feral cat faecal odour, may have been due to differences in their use of faeces as a scent mark (Dickman 1992). Rather than defecate singly along well used pathways as foxes (Macdonald 1979; Macdonald 1980) and cats do (Panaman 1981; Feldman 1994), western quolls tend to deposit faeces at communal latrines (Serena and Soderquist 1989). However, tiger quolls also use latrines (Belcher 1995; Kruuk and Jarman 1995; Burnett 2000; Claridge et al. 2004) and in this study their faecal odour was strongly avoided. Latrines may be particularly dangerous places for prey species to frequent as they represent areas visited by multiple tiger quolls several times a day (Kruuk and Jarman 1995; Claridge et al. 2004). Chapter 4: Do Australian small mammals avoid predator odours? 102

It is possible that the selective pressure for the avoidance of red fox odour by native rodents has been stronger than for avoidance of western quoll odour by house mice.

Since European settlement western quoll numbers have declined dramatically. They now occupy only 2% of their former range (Jones et al. 2003) and while they are known to prey upon house mice, rely heavily on invertebrate prey (Soderquist and

Serena 1994). In comparison, after their “successful” introduction in the 1870s, red foxes rapidly spread throughout mainland Australia (Rolls 1969), and prey heavily on native rodents, particularly where rabbits are rare or absent (Green and Osbourne 1981;

Triggs et al. 1984; Lunney et al. 1990; McKay 1994; Green 2003).

In any case, after less than 150 years of exposure native rodents avoid the odour of the red fox to the same degree as the tiger quoll with which they have co-evolved for at least a million years. It is difficult to determine if this is due to a strengthening of a generalised response to common carnivore odours or simply a rapid evolution of recognition and avoidance of fox odour, without conducting a chemical analysis to test for any common constituents in the odours of these two predators. But regardless of the mechanism, these native rodents clearly recognise and respond to the odour of the red fox in accordance with predator avoidance.

The avoidance of fox odour by all three rodent species is unexpected as Banks (1998), using similar sampling methods, reported no avoidance of fox faecal odour by bush rats in open woodland habitats in Namadgi National Park (see also Banks et al. 2003 response of bush rats to domestic dog faecal odour). One possible reason for these contrasting results is that Banks (1998) trapped for only one night, compared to the Chapter 4: Do Australian small mammals avoid predator odours? 103

three consecutive nights used in this study, and so may have only sampled the least wary or most trappable individuals in the population. Stoddart (1982) found that the trappability of individual field voles caught in traps tainted with tiger Panthera tigris urine was significantly greater than for those which were not, even though significantly fewer field voles were trapped overall in the tainted traps. Animals caught after a single night of trapping may be the most trappable, and therefore may not exhibit a level of predator odour avoidance which is representative of the entire population. To test this theory I reanalysed my results for bush rats using Binary Logistic Regression, but only included data from the first day of trapping for each site. The results were the same as when using the data from all three days. There was still no odour by design or odour by location interaction, and there was still significant avoidance of both tiger quoll and red

2 fox faecal odour (Ȥ 2=6.914, P=0.032).

Another possibility is that these contrasting results may be due to local differences in predation pressure. In the present study, bush rats were common to abundant throughout the forest and McKay (1994) reported bush rat or swamp rat remains in

44.8% of fox faeces collected from the same area. In the study by Banks (1998), although foxes were relatively common, they preyed primarily on rabbits Oryctolagus cuniculus and juvenile grey kangaroos Macropus giganteus; bush rats were restricted to vegetation along creek lines and their remains occurred in only 2.5% of fox scats

(Banks 1997). Moreover, a fox removal experiment resulted in no significant increases in bush rat numbers (Banks 1999). Thus, the predation risk posed by foxes is probably sufficiently low, such that any benefits of fox odour avoidance are outweighed by the costs in these Namadgi bush rats. Chapter 4: Do Australian small mammals avoid predator odours? 104

In this study, bush rats avoided tiger quoll odour in Garigal and Royal National Parks, where tiger quolls are now extremely rare or absent, but they were common in these areas in the past (NPWS 1998; NPWS 2000; NPWS 2005). This behaviour has been retained by bush rats, despite the lack of predation pressure, but this is a very different situation than not having evolved in the first place. Dickman (1992) found that house mice from islands where there were no mammalian predators “lost” their avoidance of red fox and feral cat faecal odour. However, on nearby Rottnest Island, where only cats were present, both predator odours were avoided equally (Dickman 1992). Although tiger quolls are now extremely rare or absent, foxes are found throughout both Garigal and Royal National Parks (NPWS 1998; NPWS 2000). Retention of tiger quoll odour avoidance may be linked to continuing avoidance of red fox odour; this could be due to common components shared by the two predator odours, but may also just be linked to the retention of this behaviour.

In contrast to the native rodent species, captures of marsupial brown antechinus were unaffected by tiger quoll or red fox odour. This mirrors the lack of response shown by long-nosed and southern brown bandicoots in Chapter 3. The lack of response by brown antechinus was not due to a low capture rate, which has hampered previous analyses of antechinus responses (Banks 1998). From the 73 antechinus caught, there was not even a trend of avoidance, yet with fewer captures a significant difference was detected for both swamp rats and eastern chestnut mice. It was also not due to the current level of predation pressure by tiger quolls or lack thereof, as the responses of both brown antechinus and bush rats were not different in the Sydney sites compared to the Myall Lakes sites. Nor was it due to the design of the experiment as there was no Chapter 4: Do Australian small mammals avoid predator odours? 105

difference in the responses of brown antechinus to the predator odours between the two sampling designs.

Dickman and Doncaster (1984) similarly found that while British rodents were trapped significantly less frequently in traps scented with fox faecal odour compared to unscented traps, the common shrew Sorex araneus entered all traps equally. Antechinus and shrews show remarkable evolutionary convergence, being small, active insectivorous mammals, common in areas with sufficient cover (Macdonald and Barrett

1993; Braithwaite 1998). Shrews are thought to be distasteful to foxes (Macdonald

1977) and Dickman and Doncaster (1984) found that while 31.3% of scats from the study area contained rodent hairs, none contained the remains of shrews. Therefore, foxes may not pose a great enough predation risk to shrews for fox-odour avoidance to have developed. In contrast, antechinus are frequently recorded in the scats of both tiger quolls and red foxes in Australia (Alexander 1980; Green and Osbourne 1981; Triggs et al. 1984; Lunney et al. 1990; McKay 1994; Belcher 1995). Dickman and Doncaster

(1984) also suggested that fresh fox droppings often attract beetles that are commonly preyed upon by shrews, and the attraction of this potential food source may outweigh any benefits from predator odour avoidance (Dickman and Doncaster 1984). This could also be the case for antechinus, as beetles are the most common food item in their diet

(Statham 1982; Fox and Archer 1984). Alternately, while neither brown antechinus nor the common shrew exhibit predator odour avoidance as explored using live trapping, other techniques may reveal that they may in fact still recognise these odours as a cue to predation risk and change their behaviour when sensing them accordingly. Chapter 4: Do Australian small mammals avoid predator odours? 106

4.7 Conclusion

This is the first study to demonstrate predator odour avoidance in native Australian mammal species. However, it appears to be restricted to the native rodents. It is possible that the ancestors of Australia’s modern day native rodent fauna evolved to avoid placental predator odours, which facilitated the evolution of comparable behaviour in response to the new marsupial predators they encountered upon entering

Australia millions of years ago. In turn, avoidance of the introduced red fox to the same degree as the native tiger quoll, after less than 150 years of predation pressure, may be due to common components in the odours of these predators, a rapidly evolved response, or a legacy of their evolutionary origin from places with placental predators.

Given the responses of bandicoots in Chapter 3 and the antechinus studied here, marsupials do not appear to avoid predator odours, but they may be changing their behaviour in response to the predation risk indicated by predator scent marks in ways which are not detectable by these trapping designs. However, if marsupials do not respond to predator odours in general, and fox odour in particular, then this may help to explain their particular vulnerability to red foxes, which have co-evolved with prey species that avoid their odour.

4.8 Acknowledgements

I would like to thank Callum Juniper, Megan Lenardon, Amy Plunkett-Cole, Alan

Russell, Tim Chapman and Jeremy Green for their help in the field. Thanks also to

Steve Henry and the CSIRO vertebrate pests unit for supplying fox faeces, and to Brad

Walker, Chad Staples and Featherdale Wildlife Park for supplying quoll faeces. I thank

Bruce Mitchell and Jennifer Kelley for their constructive comments on an earlier draft Chapter 4: Do Australian small mammals avoid predator odours? 107

of this manuscript. The research was carried out under the auspices of a NSW National

Parks and Wildlife Service (NPWS) scientific research licence B2185 and UNSW

Animal Care and Ethics Approval No. 00/64. Chapter 4: Do Australian small mammals avoid predator odours? 108

4.9 References

Alexander, D. (1980). Diet of the Tiger Quoll (Dasyurus maculatus Kerr) in an area of New England National Park, Northeastern New South Wales. Honours thesis, University of New England, Armidale.

Banks, P. B. (1997). Predator-Prey Interactions between Foxes, Rabbits and Native Mammals of the Australian Alps. Ph.D. dissertation, University of Sydney, Sydney.

Banks, P. B. (1998). Responses of Australian Bush Rats, Rattus fuscipes, to the odour of introduced Vulpes vulpes. Journal of Mammalogy 79, 1260-1264.

Banks, P. B. (1999). Predation by introduced foxes on native bush rats in Australia: do foxes take the doomed surplus? Journal of Applied Ecology. 36, 1063-1071.

Banks, P. B., Hughes, N. K. and Rose, T. A. (2003). Do native Australian small mammals avoid faeces of domestic dogs? Responses of Rattus fuscipes and Antechinus stuartii. Australian Zoologist 32, 406-409.

Belcher, C. A. (1995). Diet of the tiger quoll (Dasyurus maculatus) in East Gippsland, Victoria. Wildlife Research 22, 341-357.

Braithwaite, R. W. (1998). Brown Antechinus Antechinus stuartii. In The Mammals of Australia. (edited by R. Strahan) pp. 94-97. New Holland Publishers, Sydney.

Burnett, S. E. (2000). Ecology and Conservation Status of the northern Spot-tailed Quoll, Dasyurus maculatus: with reference to the Future of Australia's Marsupial Carnivores. Ph.D. dissertation, James Cook University of Northern Queensland, Townsville: 292.

Calder, C. J. and Gorman, M. L. (1991). The effects of Red Fox Vulpes vulpes faecal odours on the feeding behaviour of Orkney Voles Microtus arvalis. Journal of Zoology (London) 224, 599-606.

Claridge, A. W., Mifsud, G., Dawson, J. and Saxon, M. J. (2004). Use of infrared digital cameras to investigate the behaviour of cryptic species. Wildlife Research 31, 645-650.

Dickman, C. R. (1992). Predation and Habitat Shift in the House Mouse, Mus domesticus. Ecology 73, 313-322.

Dickman, C. R. and Doncaster, C. P. (1984). Responses of Small Mammals to Red Fox (Vulpes vulpes) Odour. Journal of Zoology (London) 204, 521-531.

Feldman, H. N. (1994). Methods of scent marking in the domestic cat. Canadian Journal of Zoology 72, 1093-1099.

Fox, B. J. and Archer, E. (1984). The Diets of Sminthopsis murina and Antechinus stuartii (Marsupialia: Dasyuridae) in Sympatry. Australian Wildlife Research 11, 235-48. Chapter 4: Do Australian small mammals avoid predator odours? 109

Gorman, M. L. (1980). Sweaty Mongooses and Other Smelly Carnivores. Symposium of the Zoological Society of London 45, 87-105.

Gorman, M. L. (1984). The response of prey to Stoat (Mustela erminea) scent. Journal of Zoology (London) 202, 419-423.

Gorman, M. L. and Trowbridge, B. J. (1989). The role of Odor in the Social Lives of Carnivores. In Carnivore Behavior, Ecology, and Evolution. (edited by J. L. Gittleman) pp. 57-88. Cornell University Press, New York.

Green, K. (2003). Altitudinal and temporal differences in the food of foxes (Vulpes vulpes) at alpine and subalpine altitudes in the Snowy Mountains. Wildlife Research 30, 245-253.

Green, K. and Osbourne, W. S. (1981). The diet of foxes Vulpes vulpes (L), in relation to abundance of prey above the winter snowline in New South Wales. Australian Wildlife Research 87, 239-250.

Jedrzejewski, W., Rychlik, W. and Jedrzejewska, B. (1993). Responses of bank voles to odours of seven species of predators: experimental data and their relevance to natural predator-vole relationships. Oikos 68, 251-257.

Jones, M. E., Oakwood, M., Belcher, C. A., Morris, K., Murray, A. J., Woolley, P. A., Firestone, K. B., Johnson, B. and Burnett, S. (2003). Carnivore concerns: Problems, issues and solutions for conserving Australasia's marsupial carnivores. In Predators with pouches: The biology of carnivorous marsupials. (edited by M. E. Jones, C. R. Dickman and M. Archer) pp. 422-434. CSIRO publishing, Collingwood.

Kays, R. W. and Wilson, D. E. (2002). Mammals of North America. Princeton University Press, Princeton.

Kleiman, D. (1966). Scent Marking in the Canidae. Symposium of the Zoological Society of London 18, 167-177.

Kruuk, H. and Jarman, P. J. (1995). Latrine use by the Spotted-tailed quoll (Dasyurus maculatus: Dasyuridae, Marsupialia) in its natural habitat. Journal of Zoology (London) 236, 345-349.

Lima, S. L. (1998). Nonlethal Effects in the Ecology of Predator-Prey Interactions: What are the ecological effects of anti-predator decision-making? Bioscience 48, 25-34.

Lunney, D., Triggs, B., Eby, P. and Ashby, E. (1990). Analysis of scats of dogs Canis familiaris and foxes Vulpes vulpes (Canidae: Carnivora) in coastal forests near Bega, New South Wales. Australian Wildlife Research 17, 61-68.

Macdonald, D. W. (1977). On food preferences in the Red Fox. Mammal Review 7, 7- 23. Chapter 4: Do Australian small mammals avoid predator odours? 110

Macdonald, D. W. (1979). Some Observations and Field Experiments on the Urine Marking Behaviour of the Red Fox Vulpes vulpes L. Zeitshrift fuer Tierpsychologie 51, 1-22.

Macdonald, D. W. (1980). Patterns of Scent Marking with Urine and Faeces Amongst Carnivores Communities. Symposium of the Zoological Society of London 45, 107-139.

Macdonald, D. W. and Barrett, P. (1993). Collins Field Guide: Mammals of Britain and Europe. Harper Collins, London.

McKay, G. M. (1994). Effects of introduced predators on biodiversity remnant vegetation. Australian Nature Conservation Agency, Canberra.

Mitchell, B. D. and Banks, P. B. (2005). Do wild dogs exclude foxes? Evidence for competition from dietary and spatial overlaps. Austral Ecology 30, 581-591.

Nolte, D. L., Mason, J. R., Epple, G., Arnov, E. and Campbell, D. L. (1994). Why are predator urines aversive to prey. Journal of Chemical Ecology 20, 1505-1516.

NPWS (1998). Garigal National Park Plan of Management. Hurstville, NSW National Parks and Wildlife Service: 46.

NPWS (2000). Royal National Park Plan of Management. Hurstville, NSW National Parks and Wildlife Service: 76.

NPWS (2002). Myall Lakes National Park, Little Broughton Island and Stormpetrel Nature Reserves Plan of Management. Hurstville, NSW National Parks and Wildlife Service: 148.

NPWS (2005). Atlas of NSW Wildlife, NSW National Parks and Wildlife Service. 2005.

Panaman, R. (1981). Behaviour and Ecology of Free-ranging Female Farm Cats (Felis catus L.). Zeitshrift fuer Tierpsychologie 56, 59-73.

Rolls, E. C. (1969). They All Ran Wild, the Animals and Plants that Plague Australia. Angus and Robertson, Sydney.

Rostain, R. R., Ben-David, M., Groves, P. and Randall, J. R. (2004). Why do river otters scent-mark? An experimental test of several hypotheses. Animal Behaviour 68, 703-711.

Russell, B. G. and Banks, P. B. (2005). Responses of four Critical Weight Range (CWR) marsupials to the odours of native and introduced predators. Australian Zoologist 33, 217-222.

Serena, M. and Soderquist, T. R. (1989). Spatial organization of a riparian population of the carnivorous marsupial Dasyurus geoffroii. Journal of Zoology (London) 219, 373-383. Chapter 4: Do Australian small mammals avoid predator odours? 111

Soderquist, T. R. and Serena, M. (1994). Dietary niche of the Western Quoll, Dasyurus geoffroii, in the Jarrah Forest of Western Australia. Australian Mammalogy 17, 133-136.

Statham, H. L. (1982). Antechinus stuartii (Dasyuridae: Marsupialia) diet and food availability at Petroi, northeastern New South Wales. In Carnivorous Marsupials. (edited by A. M) pp. 151-163. Royal Zoological Society of New South Wales, Sydney.

Stoddart, D. M. (1976). Effect of Odour of Weasels (Mustela nivalis L.) on Trapped Samples of Their Prey. Oecologia 22, 439-441.

Stoddart, D. M. (1982). Does trap odour influence estimation of population size of the short-tailed vole, Microtus agrestis? Journal of Animal Ecology 51, 375-386.

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

Triggs, B. (1996). Tracks, scats and other traces; A field guide to Australian mammals. Oxford University Press, South Melbourne.

Triggs, B., Brunner, H. and Cullen, J. M. (1984). The food of fox, dog and cat in Croajingalong National Park, south-eastern Victoria. Australian Wildlife Research 11, 491-499.

Wolff, J. O. and Davis-Born, R. (1997). Response of gray-tailed voles to odours of a mustelid predator: a field test. Oikos 79, 543-548. Chapter 5: Behavioural responses of small mammals to predator odours 112

Chapter 5:

Behavioural responses of Australian small mammals to the faecal odours of native and introduced predators

5.1 Abstract

Mammalian prey species may change their behaviour in response to the increased risk of predation implied by a predator’s scent mark in a number of different ways. In this chapter, I investigated whether bush rats Rattus fuscipes, swamp rats R. lutreolus and brown antechinus Antechinus stuartii altered their feeding, space use and locomotory activity in response to the faecal odours of either the native tiger quoll Dasyurus maculatus or the introduced red fox Vulpes vulpes. Wild caught individuals were placed in enclosures and their movements were videoed by overhead cameras. Changes in feeding were examined by measuring the giving up densities (GUDs) of; sunflower seeds in trays of sand for the rodents, and mealworms in woodchip for antechinus.

GUDs and movement patterns were compared between the night before and the night after predator faeces were introduced into the enclosure. Brown antechinus increased their GUDs in response to both predator odours, and when quoll faeces were used, spent less time in the enclosure where the faeces were presented. Bush rats and swamp rats responded to both predator odours by reducing their average speed of movement, presumably reducing their conspicuousness to these predators. The differences in the responses of these prey species may be due to different evolutionary histories, differences in size or different ecological requirements. The similarities in the strength of the responses of brown antechinus, bush rats and swamp rats to tiger quoll and red fox faecal odours may be due to common components in the faecal odours of these Chapter 5: Behavioural responses of small mammals to predator odours 113

predators, or these prey species may have rapidly evolved to respond to red fox odour after less than 150 years of heavy predation pressure.

5.2 Introduction

All animals are subject to the risk of predation throughout their lifetime (Endler 1991).

Those without adequate anti-predator strategies are preyed upon and removed from the gene pool, leading to the evolution of a wide range of behavioural tactics to reduce the likelihood of a successful predatory event (Lima and Dill 1990). While these behavioural modifications increase the animal’s chance of survival, there are often costs involved, including reductions in feeding or breeding success (Lima 1998). Therefore, many such behavioural modifications are often only employed under an increased risk of predation (Lima and Dill 1990) and are typically tailored to address the nature of the risk of predation, whether it be via random encounter with a predator, detection by hunting predator or an increase in predator success once detected.

Potential prey may use a number of different factors to assess this risk, and many species are known to respond to olfactory cues (Kats and Dill 1998). Mammalian predators often use scent marks as chemical signals and their odour may convey information about identity, social status, reproductive state, and territorial boundaries

(Gorman and Trowbridge 1989; Rostain et al. 2004). These scent marks are regularly revisited and remarked by the signaller and are also visited by conspecific receivers

(Kleiman 1966; Gorman 1980; Macdonald 1980; Gorman and Trowbridge 1989). They thus represent areas with an increased likelihood of encountering these predators, and may be interpreted by potential prey species as a reliable indicator of increased predation risk. Chapter 5: Behavioural responses of small mammals to predator odours 114

In Chapter 4, I found that three Australian native rodent species; the bush rat Rattus fuscipes, the swamp rat R. lutreolus, and the eastern chestnut mouse Pseudomys gracilicaudatus, avoided traps scented with the faecal odour of two predators; the native tiger quoll Dasyurus maculatus and the introduced red fox Vulpes vulpes; but the

Australian native marsupial species, the brown antechinus Antechinus stuartii, avoided neither odour. However, Powell and Banks (2004) recently suggested that this “scat at trap” technique might not be totally representative of the reactions of small mammals towards predator odours, as it only represents a snapshot of their immediate reaction, whereas many of the costs and benefits of risk avoidance strategies arise from longer- term behavioural shifts. Longer-term responses are particularly relevant to studies of olfactory cues to predation risk because the risk of predation may not be as immediate or direct as for other cues, and may only indicate a higher likelihood of predators returning. Indeed, mammalian prey species modify their behaviour in response to this increase in predation risk in a number of different ways. As discussed in Chapter 1, these behavioural modifications fall into four main categories; spatial responses, foraging responses, locomotory responses and life history responses (Ylönen 2001).

The scat at trap technique only assesses short-term spatial responses. In this chapter, I examine foraging responses, longer-term spatial responses and locomotory responses of

Australian native small mammals to olfactory cues of predation risk, to more fully address the different ways by which prey may reduce their net risk of successful predator attack.

Foraging responses of small mammals to indirect cues of predation risk, such as open space, have been well documented both in northern hemisphere rodents (Brown 1989; Chapter 5: Behavioural responses of small mammals to predator odours 115

Bowers et al. 1993; Brown et al. 1994; Brown and Morgan 1995; Jacob and Brown

2000) and small Australian dasyurid marsupials (Stokes et al. 2004). However, changes to foraging in response to olfactory cues of predation risk are more ambiguous. While larger scale studies have reported no change in the giving up densities (GUDs) of rodents in response to carnivore urine and faeces (Thorsen et al. 1998; Jones and Dayan

2000; Orrock et al. 2004; Powell and Banks 2004), smaller scale experiments have shown that a number of rodents species do reduce their feeding in the presence of mustelid and fox odours (Berdoy and Macdonald 1991; Calder and Gorman 1991;

Barreto and Macdonald 1999; Carlsen et al. 1999; Bolbroe et al. 2000; Rosell 2001).

Rather than complete avoidance, this response represents a decrease in the amount of time spent near the odour source and hence a reduction in the likelihood of predator encounter. Predation risk is still reduced, although to a lesser degree, but prey species may be willing to spend a limited amount of time close to a predator’s scent mark, until the benefits of additional food are outweighed by the increasing likelihood of encountering the predator (Brown 1988).

There have only been a few studies which have directly measured the amount of time small mammals spend in the presence of predator odours. Over a ten minute period field voles Microtus agrestis and Orkney voles M. arvalis (Gorman 1984), and meadow voles M. pennsylvanicus (Parsons and Bondrup-Nielsen 1996) spent significantly less time in areas scented with stoat Mustela erminea odour, and over 15 minutes Orkney voles spent less time in areas scented with fox faeces (Calder and Gorman 1991). The only study to consider a longer time frame found that water voles Arvicola terrestris spent significantly less time in areas scented with American mink M. vison bedding over a 21 hour period (Barreto and Macdonald 1999). As with a reduction in foraging, Chapter 5: Behavioural responses of small mammals to predator odours 116

this response does not represent complete avoidance, but does result in a reduction in the likelihood of predator encounter. Prey species may use this small amount of time spent in the vicinity of predator odour to make a rapid assessment of any potential resources, to allow for a more accurate consideration of the trade-off between acquiring these resources and the risk of predation.

Changes in locomotory activity or mobility in the presence of predator odours have also received only limited attention, but have been reported for several small mammal species, particularly from small scale lab studies. Several vole species have been shown to reduce their locomotory activity in the presence of mustelid odours (Gorman 1984;

Jedrzejewski et al. 1993; Borowski 1998b), and similarly laboratory brown rats Rattus norvegcicus reduce their locomotory activity in response to domestic cat Felis catus odour (Dielenberg et al. 2001; Dielenberg and McGregor 2001; McGregor et al. 2002).

Again, these responses do not represent complete spatial avoidance, but relate to other ways of reducing net predator risk. At a large scale, a reduction in mobility may serve to reduce the likelihood of encountering a predator (Borowski 1998b; Norrdahl and

Korpimäki 1998; Banks et al. 2000). At a smaller scale, a reduction in locomotory activity is assumed to reduce the likelihood of detection by predators, especially if the predator hunts by sight or sound (Lima and Dill 1990; Jedrzejewski et al. 1993; Kats and Dill 1998). However, a reduction in mobility may also be dangerous if it leads to increased predictability of prey location, e.g. by accumulation of prey odours (Banks et al. 2000; Banks et al. 2002).

In this chapter, I build upon the results gained in chapter 4, in order to provide a more complete understanding of odour induced anti-predator behaviour in Australian native Chapter 5: Behavioural responses of small mammals to predator odours 117

mammals. I examine these alternate strategies by which potential prey may also reduce their total exposure to predation risk, in response to olfactory cues from their predators.

Specifically, I test whether bush rats, swamp rats and brown antechinus respond to the increased risk of predation implied by tiger quoll and red fox faecal odours by changing; their feeding, the amount of time spent in the immediate vicinity of the odour source, or their locomotory activity. Wild-caught individuals were used in a captive experiment comparing the behaviour and activity patterns of these small mammals over a 24 hour period time period in the absence of predator odour to the following 24 hour period, after predator odours were presented.

5.3 Methods

5.3.1 Animals

Wild bush rats, swamp rats and brown antechinus were captured using collapsible aluminium live-capture Elliott traps in Myall Lakes National Park in New South Wales

Australia, in the areas described in Chapter 4. All animals used in the experiment were non-breeding adults: bush rats ranged in size from 70-130 g (mean 117 g); swamp rats ranged in size from 115-180 g (mean 140 g); brown antechinus ranged in size from 18-

28 g (mean 23 g).

As detailed in Chapters 3 and 4, tiger quoll faeces were collected from captive quolls held at Featherdale Wildlife Park in Sydney and red fox faeces were collected from captive foxes at the CSIRO Sustainable Ecosystems Unit in Canberra. Faeces were either used fresh, or frozen for later use. No “strong odour” control was used to test the possibility that rodents merely avoid strong novel odours because there is no odour that is strong without having unique scents associated with it (Banks 1998). Chapter 5: Behavioural responses of small mammals to predator odours 118

5.3.2 Enclosures

The experiment took place at the University of New South Wales’ Cowan Field Station

(latitude 151° 10’E, longitude 33° 35’S) in four 1.85 x 1.85 m interconnected outdoor enclosures as described by Righetti et al. (2000) (Fig. 1). The four enclosures were interconnected by openings 40cm wide and 20 cm high in the middle of each of the interior walls. The floors of the enclosures were lined with wood shavings. Enclosure 1 and 2 were designated water and shelter enclosures, enclosures 3 and 4 were designated feeding enclosures. An upturned large Elliott trap (46 x 16x 16cm) wedged open and containing non-absorbent cotton wool for bedding was placed in the far middle corner of enclosures 1 and 2 to act as a shelter/nest. Water bottles were taped to the middle of the side walls of enclosures 1 and 2. Feeding trays were placed in the centre of enclosures 3 and 4. Rat feeding trays were oval aluminium trays (45 x 30 x 7cm) filled with 1 litre of sand and 7.5 grams of sunflower kernels (125 pieces). Both species of rat have a highly variable diet which may include vegetation and insects, but seeds make up an important component of the diet at different times throughout the year (Watts and

Braithwaite 1978; Cheal 1987; Luo and Fox 1996). Antechinus feeding trays were rectangular aluminium trays (30 x 20 x 5 cm) filled with 1 litre of fine woodchip and 35 mealworms (the larval stage of the black beetle Tenebrio molitor). Coleoptera make up the major component of the diet of brown antechinus in the area (Fox and Archer 1984) and mealworms are commonly used to feed captive antechinus (Jackson 2003). At the outer end of each feeding tray several stones of a similar size, shape and colour to the predator faeces were placed on a folded sheet of absorbent paper towel. This was to control for any response to a visual change the sudden appearance of the faeces may have had. Chapter 5: Behavioural responses of small mammals to predator odours 119

Enclosure 1 Enclosure 2 Shelter

Water

Food Tray stone/ faeces Enclosure 3 Enclosure 4

Figure 5.1: Layout of interconnected experimental enclosures

Animal movements were measured using infra-red sensitive closed circuit television

(CCTV) video surveillance cameras with 12 built in infrared LEDs, which were attached to the centre of the roof of each enclosure. These cameras recorded the position of the animal within the enclosures at night without requiring additional lighting. The cameras were connected to a real time digital quad processor, which integrated the images from all four cameras into a single image, which was then recorded on a 24 hour time-lapse VCR (Mitsubishi HS-7424).

5.3.3 Trial procedure

Twelve individuals of each species (six males and six females) were tested in separate trials. Each individual was introduced into the enclosures at approximately 7:00am and Chapter 5: Behavioural responses of small mammals to predator odours 120

spent the following four days in the experiment. Every morning (7:00am), the video tapes were changed, the water bottles checked, and the remaining sunflower kernels were weighed and replenished for the rat species, while for the antechinus the remaining meal worms were counted and replenished, in order to determine the GUD which was used to measure their foraging response (sensu Brown 1988).The GUD was defined as; the weight of sunflower seeds for the rodents, and the number of mealworms for antechinus; remaining after 24 hours of foraging. After two days the stones at the end of one of the two feeding trays were replaced with approximately 60 grams of predator faeces. The animal was then left in the enclosures for another two days. At the end of each trial all materials were replaced to avoid any response by the following animal to the odour of the previous animal.

Each animal was exposed to only one of the two predator odours with equal numbers of each sex and species randomly allocated to each of the two predator odours. Animals were introduced into either enclosure 1 or enclosure 2, this was determined randomly, as was the location of the odour (either enclosure 3 or 4). When not in use in the experiment, individual animals were maintained in separate laboratory cages and were held in captivity for a maximum of six weeks. Animals were kept in the cages for a minimum of three days prior to being introduced into the experimental enclosures. The labortatory cages were lined with wood shavings, the same as the experimental enclosures, and animals were given food and water ad libitum. Bush rats were fed sunflower seeds and standard rodent pellets supplied by “D & R stockfeed”. Antechinus were fed mealworms and Wombaroo insectivore mix. Antechinus were only used in the experiment between February and May, to avoid possible behavioural shifts associated Chapter 5: Behavioural responses of small mammals to predator odours 121

with the breeding season in late May (Braithwaite 1979). After participation in the experiment all animals were returned to their place of capture.

5.3.4 Data analysis

Video footage was analysed using the computer program Ethovision 2.3 (Noldus 2002).

Data from dusk till dawn was analysed, as Ethovision requires constant illumination to track the animals successfully. Bush rats and brown antechinus are largely nocturnal

(Braithwaite 1998; Lunney 1998a) whereas swamp rats have been reported to be active both during the day, and at night (Lunney 1998b), but in the current study swamp rats were active for longer at night (mean time active per night: 9 hours 37 minutes) than either bush rats (mean time active per night: 9 hours 34 minutes) or brown antechinus

(mean time active per night: 5 hours 19 minutes). To measure spatial and locomotory responses to predator odours I used Ethovision to calculate the total time spent in each enclosure, the number of times the animal entered each of the enclosures, and average speed of movement in each of the enclosures. Values were calculated over a series of successive 30 minute time periods.

For each species, I analysed the data from the odour treated enclosure using a two- factor repeated measures analysis of variance (ANOVA). The factors were day (day 2 without odour, and day 3 with odour) and odour (fox or quoll), with the individual animals nested within odour. The four response variables tested were; GUDs, total amount of time the animals spent in the enclosure, total number of times the animals entered the enclosure, and average speed of the animal whilst in the enclosure. If there was a difference between the nights, but no interaction with odour (no difference in the response to the two predator odours) the same analyses were conducted on the data for Chapter 5: Behavioural responses of small mammals to predator odours 122

the untreated feeding enclosure to determine if changes were attributable to the predator odours or the number of nights spent in the enclosures. If the latter were the case, we would expect changes to be apparent in both the treated and untreated enclosures.

Statistical analyses were conducted using Minitab 14 (Minitab 2003). All data was tested to ensure the assumptions of ANOVA were fulfilled; normality and homogeneity of variance.

5.4 Results

5.4.1 Bush Rats

Bush rats did not alter their GUDs (Figure 5.2), the amount of time they spent in the odour-treated enclosure (Figure 5.3), or the number of times they entered the odour- treated enclosure (Figure 5.4) in response to the predator odours (Table 5.1a, b & c).

However, they did reduce their average speed of movement on day 2 compared to day 3 by an average decrease of 15.6% (Figure 5.5). There was no difference in the response to either red fox or tiger quoll faeces, as there was no day by odour interaction (Table

5.1d). Analysis of the average speed in the untreated feeding enclosure revealed a day by odour interaction (Table 5.1e). Bush rats reduced their speed in the untreated enclosure by 16.6%, when quoll odour was presented in the adjacent enclosure, but there was no change when fox odour was presented. The response was therefore not due to the number of nights spent in the enclosures, as the response was not seen in both the treated enclosure and the untreated enclosure, when fox faeces were used as the odour source. When quoll faeces were used, bush rats reduced their average speed in both enclosures, suggesting that the response to quoll faeces may be evident at greater distances from the odour source than the response to fox faeces. Chapter 5: Behavioural responses of small mammals to predator odours 123

5.4.2 Swamp Rats

The results for swamp rats were extremely similar to those for bush rats. Swamp rats did not alter their GUDs (Figure 5.2), the amount of time they spent in the odour- treated enclosure (Figure 5.3), or the number of times they entered the odour-treated enclosure (Figure 5.4) in response to the predator odours (Table 5.2a, b & c). However, as for bush rats, swamp rats significantly reduced their average speed of movement in response to both tiger quoll and red fox odour by an average decrease of 16.9% (Figure

5.5; Table 5.2d). The response was not due to the number of nights spent in the enclosures, as there was no change in the average speed of swamp rats in the untreated feeding enclosure when either predator odour was presented in the adjacent enclosure

(Table 5.2e).

5.4.3 Brown Antechinus

Brown antechinus significantly increased their GUDs in response to both tiger quoll and red fox faeces by an average of 59% (Figure 5.2; Table 5.3a). This increase in

GUDs was not due to the number of nights spent in the enclosures, as there was no change in the GUDs of brown antechinus in the non-odour enclosure (Table 5.3b).

There was an interaction between day and odour for total time spent in the odour- treated enclosure (Table 5.3c), as brown antechinus reduced the amount of time they spent in the enclosure by 39% when quoll faeces were presented, but increased the amount of time they spent in the enclosure by 29% when fox faeces were presented

(Figure 5.3). Brown antechinus did not alter the number of times they entered the odour-treated enclosure (Figure 5.4), or their average speed in this enclosure (Figure

5.5) in response to either predator odour (Table 5.3d & e). Chapter 5: Behavioural responses of small mammals to predator odours 124

2 12 fox quoll 1.5 9

1 6

0.5 3

0 0

-0.5 -3

-1 Bush Rat Swamp Rat -6 Brown Antechinus -1.5 -9 Change in GUDs (grams remaining) Brown

Change in GUDs (mealworms remaining) Antechinus -2 BushRat Swamp Rat -12

Figure 5.2: Mean ± SE change in giving up densities (GUDs) of bush rats, swamp rats and brown antechinus between day 2 (before the predator odour was introduced) and day 3 (when the odour was introduced) in the enclosure where the odour was presented.

140 fox quoll 120 100 80 60 40 20 0

change inchange (minutes) time -20 -40

-60 Bush Rat Swamp Rat Brown Antechinus

Figure 5.3: Mean ± SE change in the amount of time bush rats, swamp rats and brown antechinus spent in the enclosure where the odour was presented, between day 2 (before the odour was introduced) and day 3 (when the odour was introduced). Chapter 5: Behavioural responses of small mammals to predator odours 125

100 fox quoll

50

0

-50

-100

-150 change in the number of entrances

-200 Bush Rat Swamp Rat Brown Antechinus

Figure 5.4: Mean ± SE change in the number of times bush rats, swamp rats and brown antechinus entered (and left) the enclosure where the odour was presented, between day 2 (before the odour was introduced) and day 3 (when the odour was introduced).

1 fox quoll 0.5

0

-0.5

-1

-1.5

change in average in change speed (km/h) -2

Brown Antechinus -2.5 Bush Rat Swamp Rat

Figure 5.5: Mean ± SE change in the average speed of movement of bush rats, swamp rats and brown antechinus in the enclosure where odour was presented, between day 2 (before the odour was introduced) and day 3 (when the odour was introduced). Chapter 5: Behavioural responses of small mammals to predator odours 126

Table 5.1: Results of two-factor repeated measures ANOVAs to test whether bush rats respond to tiger quoll and red fox faecal odour by altering: (a) their GUDs in the odour treated enclosure; (b) the total amount of time they spent in the odour-treated enclosure; (c) the number of times they entered the odour treated enclosure; (d) their average speed of movement in the odour treated enclosure; (e) their average speed of movement in the untreated feeding enclosure. df F P

(a) GUDs in the odour treated enclosure odour 1 0.12 0.736 individual (odour) 10 11.64 <0.001 day 1 0.72 0.417 day by odour 1 1.29 0.283

(b) Total amount of time in the odour treated enclosure odour 1 1.75 0.216 individual (odour) 10 9.29 0.001 day 1 0.43 0.526 day by odour 1 1.52 0.256

(c) Number of times odour treated enclosure entered odour 1 0.81 0.388 individual (odour) 10 3.69 0.026 day 1 0.05 0.836 day by odour 1 1.07 0.325

(d) Average speed in the odour treated enclosure odour 1 0.07 0.803 individual (odour) 10 15.86 <0.001 day 1 5.28 0.044 day by odour 1 0.01 0.907

(c) Average speed in the untreated feeding enclosure odour 1 0.00 0.968 individual (odour) 10 13.01 <0.001 day 1 0.85 0.379 day by odour 1 5.15 0.047 Chapter 5: Behavioural responses of small mammals to predator odours 127

Table 5.2: Results of two-factor repeated measures ANOVAs to test whether swamp rats respond to tiger quoll and red fox faecal odour by altering: (a) their GUDs in the odour treated enclosure; (b) the total amount of time they spent in the odour-treated enclosure; (c) the number of times they entered the odour treated enclosure; (d) their average speed of movement in the odour treated enclosure; (e) their average speed of movement in the untreated feeding enclosure. df F P

(a) GUDs in the odour treated enclosure odour 1 0.46 0.512 individual (odour) 10 42.46 <0.001 day 1 1.23 0.294 day by odour 1 0.07 0.794

(b) Total amount of time spent in the odour treated enclosure odour 1 0.14 0.720 individual (odour) 10 0.44 0.894 day 1 0.78 0.398 day by odour 1 0.67 0.433

(c) Number of times the odour treated enclosure was entered odour 1 0.00 0.996 individual (odour) 10 5.55 0.006 day 1 1.33 0.275 day by odour 1 0.32 0.584

(d) Average speed in the odour treated enclosure odour 1 0.35 0.567 individual (odour) 10 26.78 <0.001 day 1 10.46 0.009 day by odour 1 1.06 0.329

(c) Average speed in the untreated feeding enclosure odour 1 0.28 0.606 individual (odour) 10 6.42 0.003 day 1 0.22 0.647 day by odour 1 0.08 0.778 Chapter 5: Behavioural responses of small mammals to predator odours 128

Table 5.3: Results of two-factor repeated measures ANOVAs to test whether brown antechinus respond to tiger quoll and red fox faecal odour by altering: (a) their GUDs in the odour treated enclosure; (b) their GUDs in the untreated feeding enclosure (c) the total amount of time they spent in the odour-treated enclosure; (d) the number of times they entered the odour treated enclosure; (e) their average speed of movement in the odour treated enclosure. df F P

(a) GUDs in the odour treated enclosure odour 1 0.05 0.820 individual (odour) 10 10.63 <0.001 day 1 6.26 0.031 day by odour 1 0.16 0.698

(b) GUDs in the untreated feeding enclosure odour 1 0.29 0.601 individual (odour) 10 1.89 0.166 day 1 0.17 0.689 day by odour 1 0.33 0.576

(c) Total amount of time spent in the odour treated enclosure odour 1 0.07 0.798 individual (odour) 10 5.47 0.006 day 1 0.88 0.370 day by odour 1 5.90 0.036

(d) Number of times the odour treated enclosure was entered odour 1 0.06 0.805 individual (odour) 10 20.68 <0.001 day 1 0.01 0.940 day by odour 1 2.04 0.183

(c) Average speed in the odour treated enclosure odour 1 2.71 0.131 individual (odour) 10 1.02 0.485 day 1 0.00 0.979 day by odour 1 0.31 0.591 Chapter 5: Behavioural responses of small mammals to predator odours 129

5.5 Discussion

The results of this chapter are summarised in Table 5.4. Both the native rodents and the brown antechinus altered their behaviour in response to predator odours in a manner likely to reduce their risk of predation. However, there were distinct differences between the rodents and the native marsupial species, in terms of which strategies they utilised.

Table 5.4: Summary of the responses of three species of Australian small mammals to the faecal odour of two predators, the native tiger quoll and the introduced red fox. Foraging Responses Spatial Responses Locomotory Responses Bush Rat Quoll - - p Fox - - p Swamp Rat Quoll - - p Fox - - p Brown Antechinus Quoll pp - Fox p -- p= decrease and - = no change

5.5.1 Foraging responses

Brown antechinus responded to the odour of both the tiger quoll and the red fox by increasing their GUDs. Fewer mealworms were foraged from the feeding trays after the addition of these predator odours. In contrast, neither rodent species showed any foraging trade-offs with the addition of odours from either predator. As tiger quolls and red foxes regularly return to scent marked areas (Macdonald 1979; Macdonald 1980;

Kruuk and Jarman 1995; Claridge et al. 2004), the longer potential prey spend in these Chapter 5: Behavioural responses of small mammals to predator odours 130

areas, the higher the likelihood of encountering the predators responsible for marking them. Presumably, antechinus perceived this increased risk of predation to be higher than the benefits of the acquisition of additional food.

A reduction in feeding near predator odours is a common response of small mammals and has been reported in several previous captive studies using red fox and mustelid odours (Berdoy and Macdonald 1991; Calder and Gorman 1991; Barreto and

Macdonald 1999; Carlsen et al. 1999; Bolbroe et al. 2000). Similarly, in a field experiment, Rosell (2001) found that gray squirrels Sciurus carolinensis fed less in areas scented with red fox urine and racoon Procyon lotor urine than in areas scented with human Homo sapiens urine, white-tailed deer Odocoileus virginianus urine and unscented areas. However, previous studies using GUDs to investigate the responses of rodents to predator odours have had equivocal results or no response at all (Thorsen et al. 1998; Herman and Valone 2000; Jones and Dayan 2000; Orrock et al. 2004; Powell and Banks 2004). These studies lacked an alternate odour-free feeding source in the nearby vicinity, and it is possible that these concentrated resource patches outweighed any increase in predation risk. The studies were also all either field based or in large enclosures (13 m by 13 m), used an unknown number of animals, and comparisons were made between open and closed microhabitats. These differences in experimental design may have led to an increased level of variance in the results, masking any response to the predator odours.

In this study, the lack of GUD responses shown by rodents could have arisen from experimental artefacts. For example, it is possible that sunflower kernels were a highly valued resource, such that predation risks associated with predator odours were Chapter 5: Behavioural responses of small mammals to predator odours 131

acceptable, although this resource was also available in areas without odour. It is also possible that rodents were more efficient and rapid foragers of sunflower kernels than were antechinus searching for meal worms. Short foraging times to reach satiation would mean that there was no prolonged risk of predation. Alternatively, these rodents may have changed the way they foraged which reduced predation risk but did not alter overall intake. For example, Borowski (1998a) found that root voles M. oeconomus altered their specific feeding behaviour, rather than their overall consumption, in response to stoat odour.

5.5.2 Spatial responses

Brown antechinus also spent less time in the odour treated enclosure when quoll faeces were used, but not when fox faeces were used. So although they avoided both odours at the scale of the feeding patch, the avoidance response to quoll odour occurred over a wider area. As predators sniff directly at their scent marks (Gorman 1980; Macdonald

1980), the risk of predator encounter is probably highest in the immediate vicinity of these scent marks. Avoidance at a very localised scale may serve to reduce this predation risk, however it is likely that the higher the perceived risk, the greater the distance from the scent mark at which the response will still be evident. Previous studies which have measured time spent near predator odours have either; been conducted in very small chambers measuring the amount of time spent in the scented half of the chamber as opposed to the unscented half (Gorman 1984; Calder and

Gorman 1991), or have looked at the amount of time spent in small chambers scented with predator odours compared to unscented chambers (Parsons and Bondrup-Nielsen

1996; Barreto and Macdonald 1999). This study was conducted at a larger scale, and Chapter 5: Behavioural responses of small mammals to predator odours 132

antechinus appeared to seek refuge in the unscented enclosures after predator scats were presented, presumably because these areas were perceived as safer.

5.5.3 Locomotory responses

The rodents appeared to adopt a different strategy; when presented with predator odours they did not alter their time spent near an odour source but rather changed their average speed of movement. Similar results have been reported for other murid species in response to exposure to odours from co-evolved predators. For example, laboratory brown rats showed a decrease in locomotory activity for over 2 hours after exposure to domestic cat Felis catus odour (Dielenberg et al. 2001; Dielenberg and McGregor

2001; McGregor et al. 2002). There was a reduction in mobility for bank voles

Clethrionomys glareolus exposed to red fox and weasel odour (Jedrzejewski et al.

1993), root voles decreased their locomotory activity when exposed to weasel odour

(Borowski 1998b), and field voles M. agrestis and Orkney voles M. arvalis reduced their locomotory activity in the presence of stoat odour (Gorman 1984). In each of these studies locomotory activity was measured by the number of times a line was crossed during the experimental period, but whether these results were due to a reduction in speed or an increase in time spent immobile was not quantifiably analysed. However, reduced locomotory activity in the common lizard Lacerta vivipara in response to the odour of three potential predators; the common viper Vipera berus, the smooth snake

Coronella austriaca, and the grass snake Natrix natrix; was found to be due to an increase in the amount of time common lizards spent moving in “slow motion”, a behaviour not observed in the repertoire of the species in the absence of predator odours

(Thoen et al. 1986; Van Damme et al. 1990). Chapter 5: Behavioural responses of small mammals to predator odours 133

A decrease in locomotory activity, whether due to slower movements, or more time spent immobile, may lead to a reduction in the likelihood of detection by predators, particularly if the predators rely upon visual and auditory cues in order to locate prey

(Lima and Dill 1990; Jedrzejewski et al. 1993; Kats and Dill 1998). Red foxes use such cues in the detection of their prey and can locate the position of the gnawing and rustling sounds made by small mammals to within a few degrees of their true location

(Österholm 1964). In the wild, foxes often rotate or cock their head back and forth whilst hunting, listening for these noises (Henry 1980; Henry 1996). Österholm (1964) found that in foxes hearing is the most important of the senses for near location of prey, followed by vision, and then olfaction. Henry (1996) also noted that once caught many small mammals became extremely immobile until the foxes relaxed and diverted their attention from their immobile prey, at which point the prey were able to escape. Similar studies have not been conducted for the tiger quoll, but movements and noise have been found to enhance the visual and auditory detection of prey by closely related eastern quolls D. viverrinus (Pellis and Nelson 1984) and northern quolls D. hallacatus (Pellis et al. 1992) and likely also play an important role in tiger quoll prey detection.

Therefore, a decrease in locomotory activity to reduce these visual and auditory cues is likely to be an effective anti-predator strategy against both tiger quolls and red foxes.

5.5.4 Scalar effects on locomotory responses

Difference in the nature of the response across the whole experimental set-up suggest that potential prey may be adopting different, scale dependent changes in behaviour in response to the predation risk indicated by these predator odours. Swamp rats only appeared to perceive higher risk in the enclosure where the odour source was presented.

However, bush rats also decreased their average speed in the untreated feeding Chapter 5: Behavioural responses of small mammals to predator odours 134

enclosure when tiger quoll faeces were used. As for brown antechinus, bush rats thus showed similar responses to native and introduced predator odours in the immediate vicinity of the odour source, but responded to the native predator over a wider area.

High mobility on a much larger scale has been linked with increased predation risk, possibly due to a higher likelihood of encountering a predator (Norrdahl and Korpimäki

1998; Banks et al. 2000). It is possible that these native species may change their behaviour to avoid quoll encounter over a larger scale than in the immediate vicinity of the odour source. However, from my enclosure experiments it is difficult to determine how the decrease in speed shown by bush rats and swamp rats would translate into a reduction in predator encounter on a large scale, and this behaviour is probably related to reducing the likelihood of detection by the predators.

5.5.5 Anti-predator behaviours revealed by trapping vs behavioural observations

In these behavioural trials, bush rats and swamp rats did not increase their GUDs, or decrease the amount of time they spent in the odour treated enclosure, in response to either tiger quoll or red fox faeces, yet in trapping trials, both rodent species avoided entering traps scented with these faeces (Chapter 4). Similarly, house mice avoided entering traps scented with fox faeces (Dickman 1992), but did not increase their giving up density in the presence of fox faeces (Powell and Banks 2004). Both of these techniques partly measure the duration of exposure to predation risk, albeit in different ways. However, it is also possible that lower capture rates in predator-scented traps do not directly represent avoidance of the area where the predator odour is presented.

Predator odours may induce more cautious behaviours such that rodents become less willing to enter unfamiliar enclosed spaces, such as traps. The reduction in feeding by; brown rats in response to red fox urinary odour (Berdoy and Macdonald 1991), and Chapter 5: Behavioural responses of small mammals to predator odours 135

water voles Arvicola terrestris in response to American mink M. vison odour (Barreto and Macdonald 1999), may similarly have been due to an unwillingness to enter the small feeding cages in the presence of these predator odours. Potential prey may be particularly vulnerable to predation when exiting such apparent hollows in an area with an increased likelihood of encountering predators.

In contrast, brown antechinus did not avoid entering predator-scented traps (Chapter 4), but did trade off foraging for safety by increasing their GUDs in response to both odours. They also decreased the amount of time they spent in the odour-treated enclosure when quoll faeces were used as the odour source. Parsons and Bondrup-

Nielsen (1996), similarly found that meadow voles M. pennsylvanicus did not avoid entering stoat-scented traps in the field; but in the lab spent less time in chambers scented with stoat odour. These divergent results may again be a consequence of the insensitive nature of trapping data in making inferences about animal behaviour. As

Powell and Banks (2004) point out; once trapped, an animal can no longer display any other responses to the increased risk of predation represented by the predator odour.

These animals may initially inspect these areas in order to assess the trade-off between a reduction in predation risk and the acquisition of any potential resources, but beyond this initial inspection they then spend less time in the vicinity of the odour source than they otherwise would. It is also possible that under field conditions, some wild animals are sufficiently food stressed that the risks of predation are outweighed by the need to feed and inspect novel food sources. This may be especially relevant for antechinus and its semelparous life history and sexual size dimorphism where body size is a key determinant of reproductive success. Chapter 5: Behavioural responses of small mammals to predator odours 136

5.5.6 Rodent responses vs antechinus responses

The differences in response to the predator odours between the brown antechinus and the native rodents were discussed in chapter 4. But the new results found here further highlights that anti-predator responses are contingent upon species specific factors.

Indeed, marsupial dasyurids and placental rodents appear to have developed quite different anti-predator strategies over the course of their evolution. The different responses may be a factor of size, as bush rats and swamp rats are more than five times larger than brown antechinus (Strahan 1998). Or, the differences may be due to the differences in ecology between the species. Bush rat and swamp rats are generalist omnivores, whereas the brown antechinus is insectivorous. Brown antechinus may be able to make a rapid assessment of any food resources in an area, and then leave the area once the trade-off with predation risk becomes too great. Bush rats and swamp rats on the other hand, with a wider range of potential food sources, may spend more time in an area finding and acquiring these resources, and therefore change their behaviour to avoid detection rather than encounter.

5.5.7 Conclusions

A number of key results have arisen from these experiments. It is the first demonstration of an Australian marsupial species unambiguously responding to the predation risk posed by the odour of a native predator with which they have co-evolved over millions of years. The results also show that this species, the brown antechinus, responds to the odour of the red fox, an introduced predator with which it has less than

150 years of contact, although not over as great an area as for tiger quoll odour. The native rodent species, the bush rat and the swamp rat, also responded to the faecal Chapter 5: Behavioural responses of small mammals to predator odours 137

odour of both the tiger quoll and the red fox, although with a different anti-predation strategy to the marsupial brown antechinus.

Why do the odours of such evolutionary divergent predators induce such comparable responses in their prey? It is possible that they share common volatile components which trigger a generalised response to carnivore odours(Dickman and Doncaster 1984;

Nolte et al. 1994). However, tiger quoll odour induced responses in brown antechinus and the bush rat over a greater area than did fox odour, suggesting that these prey species are able to distinguish between red fox and tiger quoll faecal odour. It is also possible that there has been rapid selection for fox-odour avoidance since the introduction of foxes less than 150 years ago, due to high predation pressure. Bush rats, swamp rats and brown antechinus are commonly preyed upon by foxes especially when rabbits are rare or absent (Green and Osbourne 1981; Lunney et al. 1990; McKay 1994;

Triggs 1996; Mitchell and Banks 2005). Banks (1999) found that the removal of red foxes had no effect on the population size of bush rats, but this does not exclude possible selective effects on bush rat behaviour. These two hypotheses could be distinguished by chemical analyses of the volatile components of fox and quoll faecal odours to test the “common constituents hypothesis”.

The results of this chapter highlight the importance of considering the different ways which prey species can respond to predator odours. Although brown antechinus did not avoid traps scented with predator faeces, they clearly do respond to olfactory cues of predation risk. Other marsupial species may similarly respond to predator odours in a manner not apparent through the “scat at trap” technique. In the following chapter, I investigate whether two other marsupial species, the northern brown bandicoot Isoodon Chapter 5: Behavioural responses of small mammals to predator odours 138

macrourus and the long-nosed bandicoot Perameles nasuta, alter their foraging, spatial or locomotory behaviour in response to tiger quoll and red fox faeces in a manner likely to reduce their risk of predation.

5.6 Acknowledgements

I would like to thank George Madani, Leanne Van Der Weyde, Michael Whitehead,

Suzi Lewis, Raquel Melendez, Brian Hawkins, Dan Ramp, Matt Hayward, Ben

Macdonald, Bruce Mitchell, Nelika Hughes, Jamie Russell, and Jonathan Russell for their help in capturing animals, thanks to Jan Nedved and Jeff Vaughn for help converting the enclosures, and thanks to Mark Russell and Diane Russell for help conducting the experiments. Thanks also to Steve Henry and the CSIRO vertebrate pests unit for supplying fox faeces, and to Brad Walker, Chad Staples and Featherdale

Wildlife Park for supplying quoll faeces. The research was carried out under the auspices of a NSW National Parks and Wildlife Service (NPWS) scientific research licence B2185 and UNSW Animal Care and Ethics Approval No. 00/64. Chapter 5: Behavioural responses of small mammals to predator odours 139

5.7 References

Banks, P. B. (1998). Responses of Australian Bush Rats, Rattus fuscipes, to the odour of introduced Vulpes vulpes. Journal of Mammalogy 79, 1260-1264.

Banks, P. B. (1999). Predation by introduced foxes on native bush rats in Australia: do foxes take the doomed surplus? Journal of Applied Ecology. 36, 1063-1071.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2000). Nonlinearity in the predation risk of prey mobility. Proceedings of the Royal Society of London: Series B, Biological Sciences. 267, 1621-1625.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2002). Mobility decisions and the predation risks of reintroduction. Biological Conservation 103, 133-138.

Barreto, G. R. and Macdonald, D. W. (1999). The response of water voles, Arvicola terrestris, to the odours of predators. Animal Behaviour 57, 1107-1112.

Berdoy, M. and Macdonald, D. W. (1991). Factors affecting feeding in wild rats. Acta OEcologica 12, 261-279.

Bolbroe, T., Jeppesen, L. L. and Leirs, H. (2000). Behavioral response of field voles under mustelid predation risk in the laboratory: more than neophobia. Annales Zooloici Fennici 37, 169-178.

Borowski, Z. (1998a). Influence of predator odour on the feeding behaviour of the root vole (Microtus oeconomus Pallas 1776). Canadian Journal of Zoology 76, 1791-1794.

Borowski, Z. (1998b). Influence of weasel (Mustela nivalis Linnaeus, 1766) odour on spatial behaviour of root voles (Microtus oeconomus Pallas, 1776). Canadian Journal of Zoology 76, 1799-1804.

Bowers, M. A., Jefferson, J. L. and Kuebler, M. G. (1993). Variation in giving-up densities of foraging chipmunks (Tamias striatus) and squirrels (Sciurus carolinensis). Oikos 66, 229-236.

Braithwaite, R. W. (1979). Social Dominance and Habitat Utilization in Antechinus stuartii (Marsupialia). Australian Journal of Zoology 27, 517-528.

Braithwaite, R. W. (1998). Brown Antechinus Antechinus stuartii. In The Mammals of Australia. (edited by R. Strahan) pp. 94-97. New Holland Publishers, Sydney.

Brown, J. S. (1988). Patch use as an indicator of habitat preference, predator risk, and competition. Behavioural Ecology and Socio-biology 22, 37-47.

Brown, J. S. (1989). Desert rodent community structure: A test of four mechanisms of coexistence. Ecological Monographs 59, 1-20.

Brown, J. S., Kotler, B. P. and Mitchell, W. A. (1994). Foraging theory, patch use, and the structure of a Negev desert granivore community. Ecology 75, 2286-2300. Chapter 5: Behavioural responses of small mammals to predator odours 140

Brown, J. S. and Morgan, R. A. (1995). Effects of foraging behaviour and spatial scale on diet selectivity: a test with fox squirrels. Oikos 74, 122-136.

Calder, C. J. and Gorman, M. L. (1991). The effects of Red Fox Vulpes vulpes faecal odours on the feeding behaviour of Orkney Voles Microtus arvalis. Journal of Zoology (London) 224, 599-606.

Carlsen, M., Lodal, J., Leirs, H. and Jensen, T. S. (1999). The effect of predation risk on body weight in the field vole, Microtus agrestis. Oikos 87, 277-285.

Cheal, D. C. (1987). The Diets and Dietary Preferences of Rattus fuscipes and Rattus lutreolus at Walkerville in Victoria. Australian Wildlife Research 14, 35-44.

Claridge, A. W., Mifsud, G., Dawson, J. and Saxon, M. J. (2004). Use of infrared digital cameras to investigate the behaviour of cryptic species. Wildlife Research 31, 645-650.

Dickman, C. R. (1992). Predation and Habitat Shift in the House Mouse, Mus domesticus. Ecology 73, 313-322.

Dickman, C. R. and Doncaster, C. P. (1984). Responses of Small Mammals to Red Fox (Vulpes vulpes) Odour. Journal of Zoology (London) 204, 521-531.

Dielenberg, R. A., Carrive, P. and McGregor, I. S. (2001). The cardiovascular and behavioral response to cat odor in rats: unconditioned and conditioned effects. Brain Research 897, 228-237.

Dielenberg, R. A. and McGregor, I. S. (2001). Defensive behavior in rats towards predatory odors: a review. Neuroscience and Biobehavioral Reviews 25, 597- 609.

Endler, J. A. (1991). Interactions between predators and prey. In Behavioural Ecology: An Evolutionary Approach. (edited by J. R. Krebs and N. B. Davies) pp. 482. Blackwell Scientific Publications, Oxford.

Fox, B. J. and Archer, E. (1984). The Diets of Sminthopsis murina and Antechinus stuartii (Marsupialia: Dasyuridae) in Sympatry. Australian Wildlife Research 11, 235-48.

Gorman, M. L. (1980). Sweaty Mongooses and Other Smelly Carnivores. Symposium of the Zoological Society of London 45, 87-105.

Gorman, M. L. (1984). The response of prey to Stoat (Mustela erminea) scent. Journal of Zoology (London) 202, 419-423.

Gorman, M. L. and Trowbridge, B. J. (1989). The role of Odor in the Social Lives of Carnivores. In Carnivore Behavior, Ecology, and Evolution. (edited by J. L. Gittleman) pp. 57-88. Cornell University Press, New York.

Green, K. and Osbourne, W. S. (1981). The diet of foxes Vulpes vulpes (L), in relation to abundance of prey above the winter snowline in New South Wales. Australian Wildlife Research 87, 239-250. Chapter 5: Behavioural responses of small mammals to predator odours 141

Henry, J. D. (1980). Fox Hunting. Natural History 89, 60-69.

Henry, J. D. (1996). Red Fox: The Catlike Canine. Smithsonian Institute Press, Washington D.C.

Herman, C. S. and Valone, T. J. (2000). The effect of mammalian predator scent on the foraging behaviour of Dipodomys merriami. Oikos 91, 139-145.

Jackson, S. (2003). Australian mammals: Biology and captive managment. CSIRO Publishing, Collingwood.

Jacob, J. and Brown, J. S. (2000). Microhabitat use, giving-up densities and temporal activity as short- and long-term anti-predator behaviors in common voles. Oikos 91, 131-138.

Jedrzejewski, W., Rychlik, W. and Jedrzejewska, B. (1993). Responses of bank voles to odours of seven species of predators: experimental data and their relevance to natural predator-vole relationships. Oikos 68, 251-257.

Jones, M. and Dayan, T. (2000). Foraging behaviour and microhabitat use by spiny mice Acomys cahirinus and A. russatus, in the presence of Blandford's fox (Vulpes vulpes) odor. Journal of Chemical Ecology 26, 455-469.

Kats, L. B. and Dill, L. M. (1998). The scent of death: Chemosensory assessment of predation risk by prey animals. Ecoscience 5, 361-394.

Kleiman, D. (1966). Scent Marking in the Canidae. Symposium of the Zoological Society of London 18, 167-177.

Kruuk, H. and Jarman, P. J. (1995). Latrine use by the Spotted-tailed quoll (Dasyurus maculatus: Dasyuridae, Marsupialia) in its natural habitat. Journal of Zoology (London) 236, 345-349.

Lima, S. L. (1998). Nonlethal Effects in the Ecology of Predator-Prey Interactions: What are the ecological effects of anti-predator decision-making? Bioscience 48, 25-34.

Lima, S. L. and Dill, L. M. (1990). Behavioral decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology 68, 619-640.

Lunney, D. (1998a). Bush Rat Rattus fuscipes. In The Mammals of Australia. (edited by R. Strahan) pp. 651-653. New Holland Publishers, Sydney.

Lunney, D. (1998b). Swamp Rat Rattus lutreolus. In The Mammals of Australia. (edited by R. Strahan) pp. 655-657. New Holland Publishers, Sydney.

Lunney, D., Triggs, B., Eby, P. and Ashby, E. (1990). Analysis of scats of dogs Canis familiaris and foxes Vulpes vulpes (Canidae: Carnivora) in coastal forests near Bega, New South Wales. Australian Wildlife Research 17, 61-68. Chapter 5: Behavioural responses of small mammals to predator odours 142

Luo, J. and Fox, B. J. (1996). Seasonal and successional dietary shifts of two sympatric rodents in coastal heathland: A possible mechanism for coexistence. Australian Journal of Ecology 21, 121-132.

Macdonald, D. W. (1979). Some Observations and Field Experiments on the Urine Marking Behaviour of the Red Fox Vulpes vulpes L. Zeitshrift fuer Tierpsychologie 51, 1-22.

Macdonald, D. W. (1980). Patterns of Scent Marking with Urine and Faeces Amongst Carnivores Communities. Symposium of the Zoological Society of London 45, 107-139.

McGregor, I. S., Schram, L., Ambermoon, P. and Dielenberg, R. A. (2002). Not all 'predator odours' are equal: cat odour but not 2,3,5 trimethylthiazoline (TMT; fox odour) elicits specific defensive behaviour in rats. Behavioural Brain Research 129, 1-16.

McKay, G. M. (1994). Effects of introduced predators on biodiversity remnant vegetation. Australian Nature Conservation Agency, Canberra.

Minitab (2003). Minitab Release 14.1, Statistical Software. Pennsylvania, Minitab.

Mitchell, B. D. and Banks, P. B. (2005). Do wild dogs exclude foxes? Evidence for competition from dietary and spatial overlaps. Austral Ecology 30, 581-591.

Noldus (2002). Ethovision 2.3, Video Tracking system for automation of behavioural experiments. Wageningen, Noldus Information Technology.

Nolte, D. L., Mason, J. R., Epple, G., Arnov, E. and Campbell, D. L. (1994). Why are predator urines aversive to prey. Journal of Chemical Ecology 20, 1505-1516.

Norrdahl, K. and Korpimäki, E. (1998). Does mobility or sex of voles affect risk of predation by mammalian predators? Ecology 79, 433-437.

Orrock, J. L., Danielson, B. J. and Brinkerhoff, R. J. (2004). Rodent foraging is affected by indirect, but not direct, cues of predation risk. Behavioural Ecology 15, 433- 437.

Österholm, H. (1964). The significance of distance receptors in the feeding behaviour of the fox (Vulpes vulpes L.). Acta Zoologica Fennica 106, 1-31.

Parsons, G. J. and Bondrup-Nielsen, S. (1996). Experimental analysis of behaviour of meadow voles (Microtus pennsylvanicus) to odours of the short-tailed weasel (Mustela erminea). Ecoscience 3, 63-69.

Pellis, S. M. and Nelson, J. E. (1984). Some aspects of the predatory behaviour of the Quoll, Dasyurus viverrinus (Marsupialia: Dasyuridae). Australian Mammalogy 15, 55-60.

Pellis, S. M., Vanderlely, R. and Nelson, J. E. (1992). The roles of vision and vibrissae in the predatory behaviour of Northern Quolls Dasyurus hallacatus (Marsupialia: Dasyuridae). Australian Mammalogy 15, 55-60. Chapter 5: Behavioural responses of small mammals to predator odours 143

Powell, F. and Banks, P. B. (2004). Do house mice modify their foraging behaviour in response to predator odours and habitat? Animal Behaviour 67, 753-759.

Righetti, J., Fox, B. J. and Croft, D. B. (2000). Behavioural mechanisms of competition in small dasyurid marsupials. Australian Journal of Zoology 48, 561-576.

Rosell, F. (2001). Effectiveness of predator odors as gray squirrel repellents. Canadian Journal of Zoology 79, 1719-1723.

Rostain, R. R., Ben-David, M., Groves, P. and Randall, J. R. (2004). Why do river otters scent-mark? An experimental test of several hypotheses. Animal Behaviour 68, 703-711.

Stokes, V. L., Pech, R. P., Banks, P. B. and Arthur, A. D. (2004). Foraging behaviour and habitat use by Antechinus flavipes and Sminthopsis murina (Marsupialia: Dasyuridae) in response to predation risk in eucalypt woodland. Biological- Conservation 117, 331-342.

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

Thoen, C., Bauwens, D. and Verheyen, R. F. (1986). Chemoreceptive and behavioural responses of the common lizard Lacerta vivipara to snaked chemical deposits. Animal Behaviour 34, 1805-1813.

Thorsen, J. M., Morgan, R. A., Brown, J. S. and Norman, J. E. (1998). Direct and indirect cues of predatory risk and patch use by fox squirrels and thirteen-lined ground squirrels. Behavioural Ecology 9, 151-157.

Triggs, B. (1996). Tracks, scats and other traces; A field guide to Australian mammals. Oxford University Press, South Melbourne.

Van Damme, R., Bauwens, D., Vandertighelen, D. and Verheyen, R. F. (1990). Responses of the lizard Lacerta vivipara to predator chemical cues: the effects of temperature. Animal Behaviour 40, 298-305.

Watts, C. H. S. and Braithwaite, R. W. (1978). The Diet of Rattus lutreolus and Five other rodents in Southern Victoria. Australian Wildlife Research 5, 47-57.

Ylönen, H. (2001). Predator Odours and Behavioural Responses of Rodents: An Evolutionary Perspective. In Advances in Pest Management 2. (edited by H. J. Pelz, D. P. Cowan and C. J. Feare). Filander Verlag, Furth. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 144

Chapter 6:

Anti-predator strategies of Australian bandicoot species in response to olfactory cues from native and introduced predators

6.1 Abstract

It has been argued that red foxes Vulpes vulpes have had such a dramatic impact on

Australia’s mammalian fauna because the anti-predator strategies of native prey species are inadequate to cope with such an efficient predator. However, in the northern hemisphere many anti-predator strategies are only initiated under increased predation risk, as assessed by olfactory cues from the predators. In this chapter, I investigated whether long-nosed bandicoots Perameles nasuta and northern brown bandicoots

Isoodon macrourus responded to the faecal odour of the red fox and a native predator the tiger quoll Dasyurus maculatus in a manner likely to reduce their risk of predation.

Each animal was placed in an enclosure and their movement patterns were videoed by overhead cameras to examine changes in behaviour between the night before and the night after predator faeces were introduced. Long-nosed bandicoots and northern brown bandicoots both increased their giving up densities after the faeces of either the tiger quoll or the red fox were introduced. In response to tiger quoll odour, both bandicoot species also decreased the amount of time they spent in the odour-treated pen. This response was stronger for long-nosed bandicoots than northern brown bandicoots, but northern brown bandicoots also reduced their average speed of movement after the faeces of either predator were introduced. The results suggest that these bandicoot species do have anti-predator strategies initiated by the olfactory cues of predators, but Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 145

also suggest that they respond more strongly to the odour of the tiger quoll than the red fox. Native prey species may still be evolving to recognise the threat of predation represented by the red fox, and therefore remain vulnerable to heavy predation.

6.2 Introduction

Complex strategies of predator avoidance are essential for the survival of any potential prey being hunted by behaviourally complex predators. Considerable work has now shown that when performing essential social and foraging activities, prey will adopt tactics to reduce risks of avoid encountering predator, particularly avoidance of areas where the likelihood of predator visitation is high (Caro 2005). However, these tactics are unlikely to be sufficient to prevent detection by actively foraging predators, which themselves employ a range of strategies to find cryptic prey (Curio 1976). As the costs of anti-predator strategies differ for different species depending upon their life style, it is unlikely that all potential prey will respond to predation risk in the same manner. In

Chapter 5, I showed that native Australian small mammals adopt several strategies of avoidance and detection in response to perceived predation risk from native and introduced predators, beyond the simple avoidance I found in Chapter 4. In this chapter,

I take a similar approach for the medium-sized bandicoots, to investigate whether they have other anti-predator tactics, beyond those investigated by trapping in Chapter 3.

The bilbies and bandicoots of the family Peramelidae have declined more than any other group of Australian marsupials (Lyne 1990). Of eleven Australian species, only two species of peramelid, the northern brown bandicoot Isoodon macrourus and the long-nosed bandicoot Perameles nasuta remain widespread and common (Strahan

1998). Three species, the pig-footed bandicoot Chaeropus ecaudatus, the desert Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 146

bandicoot P. eremiana and the lesser bilby Macrotis leucura are now extinct, and another five are listed as either endangered or vulnerable to extinction (Maxwell et al.

1996). Of these five species, the arrival of the introduced red fox Vulpes vulpes is known to have resulted in mass predation and extinction of populations of southern brown bandicoots I. obesulus (Preiss and Kraehenbuehl 1966; Hornsby 1984; Paull

1995) and eastern barred bandicoots P. gunnii (Seebeck 1979; Seebeck 1990) and has been linked to the pattern of decline of the western barred bandicoot P. bougainville

(Richards and Short 2003) and the greater bilby M. lagotis (Watts 1969; Southgate

1990). Kinnear et al. (2002) suggested that one reason why red foxes had such a significant impact on Australia’s native fauna was because the anti-predator strategies of native prey species, while sufficient to cope with native predators, were ineffective against a predator as efficient as the red fox. Yet the strategies used to avoid predation by similar native predators such as tiger quolls Dasyurus maculatus are poorly known, and may only be implemented under periods of perceived increased predation risk

(Lima and Dill 1990).

In chapter 3, northern brown bandicoots appeared to respond to the scent marks of the tiger quoll, as they were captured more often in traps scented with tiger quoll faeces than unscented traps, but were unaffected by red fox faecal odour. Captures of long- nosed bandicoots were unaffected by the faecal odour of either predator. However, these bandicoots may respond to predator odours in ways unrelated to trappability. For example, common brushtail possums Trichosurus vulpecula have previously been found to have a higher giving up density (GUD) in the presence of red fox urine and faeces (Gresser 1996), yet captures of common brushtail possums were unaffected by red fox faecal odour in chapter 3. Similarly, in chapter 5 brown antechinus Antechinus Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 147

stuartii responded to both tiger quoll and red fox faecal odour by increasing their

GUDs, and spent less time in the pen where quoll odour was presented, despite captures of antechinus being unaffected by either predator odour in chapter 4.

Most work examining the responses of medium-sized mammals to predator odours has concentrated on the development of odour-based browsing repellents in northern hemisphere systems. However, although the faeces and urine of sympatric predators reduces browsing by snowshoe hares Lepus americanus (Sullivan et al. 1985), woodchucks Marmota monax (Swihart 1991), and mountain beavers Aplodontia rufa

(Epple et al. 1993), little is known about the underlying behavioural mechanisms producing this browsing reduction. It is possible that many of the anti-predator behaviours of small mammals, such as changes in feeding or movement, may also be utilised by medium sized mammals. For example, Western hedgehogs Erinaceus europaeus, which are ecologically quite similar to bandicoots, reduce their feeding in the presence of Eurasian badger Meles meles faeces in both the lab and field, but in the field, badger odours also cause hedgehogs to cross their own tracks less often, sniff the air more often, and alter their average speed of movement (Ward et al. 1997). On the other hand, small mammal strategies of freezing, creeping or crypsis under intense predation risk might not be viable for larger prey, which are inherently more conspicuous and may not be able to rely upon concealment.

In this chapter, I examine whether northern brown bandicoots and long-nosed bandicoots respond to tiger quoll and red fox faecal odours by reducing their feeding, the amount of time spent in the immediate vicinity of the odour source, or their average speed of movement. As in chapter 5, wild-caught individuals were used in a captive Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 148

experiment comparing behaviour and activity patterns over a 24 hour time period in the absence of predator odour to the following 24 hour period, after either tiger quoll or red fox faeces were introduced.

6.3 Methods

6.3.1 Animals

Wild northern brown bandicoots were captured using large Elliott traps (46 x 16 x 16 cm) from areas in Myall Lakes National Park, as described in Chapter 3. Long-nosed bandicoots were captured from areas of closed woodland within Muogamarra Nature

Reserve (latitude 33° 35’S, longitude 151° 10’E) in Sydney’s north, using cage traps

(50 x 19 x 19cm) wrapped in hessian to provide protection from the elements, and baited with a mixture of rolled oats, peanut butter and vegetable oil. Only adult animals were used in the experiment: northern brown bandicoots ranged in size from 900-2100 g (mean 1625 g); long-nosed bandicoots ranged in size from 600-1200 g (mean 975 g).

Both bandicoot species are preyed upon by quolls (Alexander 1980; Belcher 1995;

Jones and Barmuta 1998) and foxes (McKay 1994; Mitchell and Banks 2005). As in

Chapters 3, 4 and 5, tiger quoll faeces were collected from captive quolls held at

Featherdale Wildlife Park in Sydney and red fox faeces were collected from captive foxes at the CSIRO Sustainable Ecosystems Unit in Canberra.

6.3.2 Enclosure

The experiment took place at the University of New South Wales’ Cowan Field Station located within Muogamarra Nature Reserve. An outdoor enclosure 10.3 x 6.7 m surrounded by a wire mesh fence 1.83 m high was divided into four 5.15 x 3.35 m pens with sheets of corrugated iron buried 40 cm below and standing 1.1 m above the ground Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 149

(Figure 6.1). Only three of these pens were used in the experiment, forming an L shape.

Access between the pens was provided by 55 cm gaps in the centre of the corrugated iron divisions. A base of woodshavings was spread over the soil to enhance contrast for videoing. Water and straw for nesting material were supplied in the central pen. In the centre of each peripheral pen was a plastic half cylindrical shelter 36 cm high and 75 cm long facing the opposite direction to the access from the central pen. Within each shelter was an oval aluminium tray (45 x 30 x 7 cm) containing 143 g of small kibble dry dog food, mixed in 2 L of sand. At the base of the feeding tray I placed several faeces-shaped stones upon a folded sheet of absorbent paper to control for any visual changes caused by the addition of the faeces.

Infra-red sensitive closed circuit television (CCTV) video surveillance cameras with 12 built in infrared LEDs were attached to a frame work suspended above the centre of each pen. As in Chapter 5, the cameras were connected to a real time digital quad processor, which integrated the images from all three cameras into a single image, which was then recorded on a 24 hour time-lapse VCR (Mitsubishi HS-7424).

While not in use in the experiment, animals were maintained within the field station in naturally vegetated yards the same size and structure as the experimental enclosure. The bandicoots were kept in these yards for a minimum of three days before being introduced to the experimental eclosure. They were fed daily on a diet of fruit, bread and dry dog food, and also foraged naturally within the yards for invertebrates. After participation in the experiment, all animals were returned to their place of capture and released. They were held in captivity for a maximum of six weeks. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 150

Pen 1 Pen 2

straw stone/ faeces water

Food shelter

Pen 3

Figure 6.1: Plan of experimental enclosure

6.3.3 Trial procedure

Eight individuals (four male and four female) of each species were tested in separate trials. Each animal was exposed to only one of the two predator odours. Animals were introduced into the central pen at approximately 7:00am, and spent four full days in the experiment. Each morning (7:00am) the video tapes were changed, the water checked, and the remaining dog food weighed and replaced, in order to determine the GUD

(sensu Brown 1988). The GUD was defined as the weight of dry dog food remaining after 24 hours of foraging. After two days the stones at the end of one of the two feeding trays were replaced with approximately 60 g of predator faeces. The animal Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 151

was left in the enclosure for another two days. The odour to which they were exposed was determined randomly, as was the location of the odour (either of the two peripheral pens).

6.3.4 Data analysis

Ethovision 2.3 (Noldus 2002) was used to analyse the video footage from dusk till dawn, as Ethovision requires constant lighting to track the animals successfully. Both bandicoot species are largely nocturnal (Stodart 1966; Gordon 1974). As in chapter 5,

Ethovision was used to calculate the total time spent in each pen, the number of times the animal entered each pen, and average speed of movement in each pen and values were calculated over a series of successive 30 minute time periods. Only periods when the animal was active were included in the analyses.

Bandicoot behavioural responses were analysed in a three-factor repeated measures

ANOVA with species (northern brown bandicoots or long-nosed bandicoot), day (day 2 without odour, and day 3 with odour) and odour (fox or quoll) as the factors, and individual animals nested within odour and species. The response variables tested were; the GUDs, total amount of time the animal spent active in the pen, total number of times the animals entered the pen, and average speed of the animal in the odour-treated pen. As in chapter 5, if there was a difference between the nights, but no interaction with odour (no difference in the response to the two predator odours), the same analyses were conducted on the data for the untreated feeding pen, to determine if changes were attributable to the predator odours or the number of nights spent in the enclosure. If the latter were the case, we would expect changes to be apparent in both Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 152

the treated and untreated pens. Statistical analyses were conducted using Minitab 14

(Minitab 2003).

6.4 Results

6.4.1 GUDs

Northern brown bandicoots and long-nosed bandicoots both foraged significantly less from the feeding trays in the odour-treated pen after either tiger quoll or red fox faeces were introduced (Table 6.1a; Figure 6.2a). This increase in GUDs was not due to the number of nights spent in the enclosure, as although there was also a significant difference in GUDs between day 2 and day 3 in the untreated feeding pen (Table 6.1b), it was in the opposite direction to the odour-treated pen. Northern brown bandicoots foraged more from the feeding trays in the untreated pen after both fox and quoll faeces were presented in the odour-treated pen: long-nosed bandicoots foraged more in the untreated pen after quoll odour was presented in the odour-treated pen, and the same amount of food in the untreated pen after fox odour was presented in the odour-treated pen (Figure 6.2b).

6.4.2 Time spent in the odour-treated pen

In terms of the amount of time spent in the odour-treated pen, there was a day by odour interaction and a day by species interaction (Table 6.2). Both bandicoot species spent less time in the odour-treated pen after tiger quoll faeces were presented, although the reduction in time was greater for long-nosed bandicoots (57%) than northern brown bandicoots (31%). After fox faeces were presented, long-nosed bandicoots did not change the amount of time they spent in the odour-treated pen, and northern brown Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 153

bandicoots increased the amount of time they spent in the odour-treated pen by 15%

(Figure 6.3).

6.4.3 Number of times the odour-treated pen was entered

There was also a significant day by odour interaction, a significant day by species interaction and a marginally significant (p=0.067) day by odour by species interaction for the number of times the odour-treated pen was entered (Table 6.3). Although neither bandicoot species changed the number of times they entered the odour-treated pen after fox faeces were introduced, both species entered the odour-treated pen less often after quoll faeces were added, and this response was greater for long-nosed bandicoots

(60%) than northern brown bandicoots (32%) (Figure 6.4).

6.4.4 Average speed

Although northern brown bandicoots consistently reduced their average speed in the odour-treated pen to a similar degree in response to both predator odours(20%), the

ANOVA did not reveal a significant result due to the high level of variation in the response of long-nosed bandicoots, in particular to quoll odour (Table 6.4; Figure 6.5).

Half of the long-nosed bandicoots exposed to quoll odour decreased their speed to a similar degree as the northern brown bandicoots, but the other half increased their speed to the same degree. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 154

Table 6.1: Results of three-factor repeated measures ANOVAs to test whether northern brown bandicoots and long-nosed bandicoots altered: (a) their GUDs in the odour- treated pen; (b) their GUDs in the untreated feeding pen. df F P

(a) GUDs in the odour-treated pen odour 1 0.05 0.832 species 1 2.47 0.142 individual (odour, species) 12 8.60 <0.001 day 1 4.87 0.048 odour by day 1 0.00 0.982 species by day 1 0.12 0.739 odour by species 1 1.47 0.249 odour by species by day 1 1.23 0.289

(b) GUDs in the untreated feeding pen odour 1 0.04 0.838 species 1 3.10 0.104 individual (odour, species) 12 11.36 <0.001 day 1 9.52 0.009 odour by day 1 0.09 0.768 species by day 1 3.92 0.071 odour by species 1 0.64 0.438 odour by species by day 1 0.75 0.404 Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 155

(a) odour treated pen

40 Northern Brown Bandicoot Long-nosed Bandicoot 35 30 25 20 15 10 5 0

Change in GUDs in Change remaining) (grams fox quoll -5

(b) untreated feeding pen

5 Northern Brown Bandicoot Long-nosed Bandicoot 0 -5 -10 -15 -20 -25 -30 -35 fox quoll Change in GUDs in Change remaining) (grams -40

Figure 6.2: Mean ± SE change in giving up densities (GUDs) of northern brown bandicoots and long-nosed bandicoots between day 2 (before the odour was introduced) to day 3 (after the odour was introduced) in: (a) the pen where the odour was presented; and (b) the untreated feeding pen. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 156

Table 6.2: Results of three-factor repeated measures ANOVAs to test whether northern brown bandicoots and long-nosed bandicoots altered the total amount of time they spent in the odour-treated pen. df F P odour 1 0.97 0.345 species 1 0.84 0.377 individual (odour, species) 12 6.84 0.001 day 1 2.44 0.144 odour by day 1 7.49 0.018 species by day 1 4.81 0.049 odour by species 1 1.44 0.253 odour by species by day 1 0.18 0.682

100 Northern Brown Bandicoot Long-nosed Bandicoot 80 60 40 20 0 -20 -40 -60 -80 -100

Change in time spent active (minutes) fox quoll -120

Figure 6.3: Mean ± SE change between day 2 (before the odour was introduced) and day 3 (after the odour was introduced) in the amount of time northern brown bandicoots and long-nosed bandicoots spent in the pen where the odour was presented. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 157

Table 6.3: Results of three-factor repeated measures ANOVAs to test whether northern brown bandicoots and long-nosed bandicoots altered the total number of times they entered the odour-treated pen. df F P odour 1 0.19 0.670 species 1 5.63 0.035 individual (odour, species) 12 9.35 <0.001 day 1 11.11 0.006 odour by day 1 12.77 0.004 species by day 1 7.72 0.017 odour by species 1 0.06 0.808 odour by species by day 1 3.85 0.073

30 Northern Brown Bandicoot Long-nosed Bandicoot 20 10 0 -10 -20 -30 -40 -50 -60 -70 fox quoll

change in number of times pen entered -80

Figure 6.4: Mean ± SE change between day 2 (before the odour was introduced) and day 3 (after the odour was introduced) in the number of times northern brown bandicoots and long-nosed bandicoots entered the pen where the odour was presented. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 158

Table 6.4: Results of three-factor repeated measures ANOVAs to test whether northern brown bandicoots and long-nosed bandicoots altered their average speed in the odour- treated pen. df F P odour 1 0.13 0.723 species 1 10.17 0.008 individual (odour, species) 12 5.62 0.003 day 1 1.33 0.271 odour by day 1 0.09 0.766 species by day 1 0.08 0.787 odour by species 1 0.16 0.699 odour by species by day 1 0.09 0.766

1 Northern Brown Bandicoot Long-nosed Bandicoot

0.5

0

-0.5

-1

-1.5

-2

change in average in change speed (km/h) fox quoll -2.5

Figure 6.5: Mean ± SE change between day 2 (before the odour was introduced) and day 3 (after the odour was introduced) in the average speed of northern brown bandicoots and long-nosed bandicoots in the pen where the odour was presented. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 159

6.5 Discussion

A closer examination of bandicoot foraging and movement responses to predator odours revealed a greater repertoire of avoidance strategies than was revealed by trapping alone (Chapter 3). Although sample sizes were small and variance sometimes large, there were clear patterns in the nature of the bandicoot responses. In summary, both bandicoot species decreased their foraging in response to either predator odour, and both species also decreased the amount of time they spent in, and the number of times they entered the odour-treated yard, when quoll faeces were used. Although these spatial responses were stronger for long-nosed bandicoots than northern brown bandicoots, northern brown bandicoots also decreased their average speed in response to both fox and quoll odour (Table 6.5).

Table 6.5: Summary of the responses of northern brown bandicoots and long-nosed bandicoots to the faecal odour of two predators; the native tiger quoll and the introduced red fox. Foraging Time in Entrances into Average odour- odour-treated speed treated pen pen Northern Brown Bandicoot quoll pp p p fox pn - p Long-nosed Bandicoot quoll pp p - fox p --- p = decrease, n = increase and - = no change Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 160

6.5.1 Foraging responses

Both species of bandicoots decreased foraging (increased their GUDs) in the odour- treated pen after either tiger quoll or red fox faeces were introduced. By foraging less food from the feeding trays in the odour-treated pen, the bandicoots presumably decreased the amount of time they spent in close proximity to the predator faeces. As tiger quolls and red foxes regularly return to scent-marked areas (Macdonald 1979;

Macdonald 1980; Kruuk and Jarman 1995; Claridge et al. 2004), a reduction in time spent foraging near the faeces of these predators in the wild should reduce their likelihood of predator encounter.

The extent to which this behaviour is displayed may be dependent on the availability of other “safe” food resources. In this study, bandicoots generally increased their feeding from the untreated feeding tray to a similar degree to which they decreased their feeding in the odour-treated pen after predator faeces were introduced. However, previous studies using GUDs to investigate the response of rodent species, where an identical odour-free alternative foraging source was not provided in close proximity, have had equivocal results or no response at all (Thorsen et al. 1998; Herman and

Valone 2000; Jones and Dayan 2000; Orrock et al. 2004; Powell and Banks 2004). The feeding trays in these GUD experiments represent resource rich feeding patches (Brown

1988). If alternate feeding sources are insufficient or sparse, it might be beneficial for the animal to maximally forage from these feeding trays, which are more concentrated with high quality food than the surrounding area. Potential prey could meet their nightly foraging requirements from these resource rich patches more quickly, and spend less time overall in the scented area, thereby reducing their risk of predation. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 161

6.5.2 Movement responses

For both bandicoots, the time spent in the odour-treated pen was also reduced by quoll odour, but not by fox odour. This behaviour was associated with an equivalent decrease in the number of times the odour-treated pen was entered suggesting that bandicoots were more reluctant to enter the “risky” pens smelling of quoll. Thus, the bandicoots responded to the two predators odours similarly at the scale of the feeding tray, but they avoided the odour of the native tiger quoll over a wider area than the odour of the introduced red fox (see below and also Chapter 5). By avoiding this larger area, they presumably further reduce the likelihood of encountering the tiger quolls responsible for depositing the faeces. Movement responses have not been reported for other medium-sized mammals, but a number of small mammal species have been shown to spend less time in areas tainted with predator odours compared to untainted areas

(Gorman 1984; Calder and Gorman 1991; Parsons and Bondrup-Nielsen 1996; Barreto and Macdonald 1999).

Differences in movement strategies between the two bandicoot species were also apparent. Long-nosed bandicoots appeared to respond more strongly to quoll odour than northern brown bandicoots in terms of the decrease in time spent in the treated pen. However northern brown bandicoots also reduced their mobility in response to both predator odours by reducing their speed by ca. 20%, whereas the speed of long nosed bandicoots was unaffected. As discussed in chapter 5, such a reduction in speed may reduce the likelihood of detection by predators, especially if the predators rely on visual and auditory cues to locate their prey (Lima and Dill 1990; Jedrzejewski et al.

1993; Kats and Dill 1998). Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 162

These differential responses may be due to differences in habitat preferences between the two bandicoot species. Long-nosed bandicoots nest in dense undergrowth, but frequently traverse and forage in more open environments (Chambers and Dickman

2002); in contrast, northern brown bandicoots tend to be restricted to thicker vegetation

(Gordon 1974; Claridge et al. 1991). In more open environments, potential prey are more conspicuous to predators (Dickman 1992), and the long-nosed bandicoot strategy of avoidance of predators may be a useful tactic to decrease the likelihood of encounter.

However, as northern brown bandicoots tend to stay in more closed environments, a reduction in movement when predation risk is high may be useful in maintaining crypsis and reducing detection by predators. These results support the conclusions of chapter 3, that northern brown bandicoots were caught more often in quoll-scented traps than unscented traps because they opt for a strategy of concealment, rather than flight, in response to predation risk. Further evidence that the two species adopt different anti-predator strategies comes from their reactions to the disturbance of their nests. When long-nosed bandicoot nests were approached in the holding yard, in order to remove an animal and place it in the experimental enclosure, all eight long-nosed bandicoots rapidly exited their nests at the slightest disturbance, presumably to escape from any potential predators. The northern brown bandicoots on the other hand, did not move even when touched, unless their nest was completely uncovered.

6.5.3 Conservation implications

Both northern brown bandicoots and long-nosed bandicoots clearly responded to the faecal odour of the tiger quoll, a predator with which they coevolved over millions of years (Strahan 1998). However, the odour of foxes, an introduced predator which they have been in contact with for less than 150 years (Rolls 1969), also affected bandicoot Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 163

foraging. As for the small mammal responses reported in chapter 5, it is possible that this was due to a generalised response triggered by volatile components common to the odours of tiger quolls and red foxes (Dickman and Doncaster 1984) derived from a carnivorous diet (Nolte et al. 1994). Alternately, they may have begun to evolve to respond to red fox faecal odour after less than 150 years of exposure. Again, chemical analyses of the volatile components of these odours are needed before this issue can be resolved.

Regardless, the results here, together with those from chapter 5, show that these marsupial prey species do not respond to red fox odour to the same degree as tiger quoll odour, despite the fact that foxes now present the greater predation risk. Northern brown bandicoots, long-nosed bandicoots and brown antechinus are all commonly preyed upon by both red foxes (Green and Osbourne 1981; Lunney et al. 1990; McKay

1994; Triggs 1996; Mitchell and Banks 2005) and tiger quolls (Alexander 1980;

Belcher 1995; Jones and Barmuta 1998), but the red fox is considered abundant throughout non-tropical mainland Australia (Coman 1998), while the tiger quoll is considered uncommon to rare on the Australian mainland (Edgar and Belcher 1998).

Short et al. (2002) stated that inchoate predator/prey systems may persist or collapse based on the balance between the hunting skills of the predator and the anti-predator defences of the prey, and suggested that surplus killings indicated that native prey were extremely vulnerable to a predators as efficient as the foxes. But predator/prey relationships, inchoate or otherwise, are dynamic, as predator and prey are involved in an evolutionary arms race of attack and defence (Dawkins and Krebs 1979). My results, using species which have survived the invasion of foxes throughout Australia, suggest that some native prey species are beginning to evolve to recognise the threat of Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 164

predation presented by foxes. However, while this process is ongoing many species remain vulnerable to heavy predation by these introduced predators.

6.6 Acknowledgements

I would like to thank George Madani, Leanne Van Der Weyde, Michael Whitehead,

Suzi Lewis, Raquel Melendez, Brian Hawkins, Dan Ramp, Matt Hayward, Ben

Macdonald, Bruce Mitchell, Nelika Hughes, Jamie Russell, and Jonathan Russell for their help in capturing bandicoots, thanks to Jan Nedved and Jeff Vaughn for help converting the enclosure, and thanks to Diane Russell and Mark Russell for help conducting the experiment. Thanks also to Steve Henry and the CSIRO vertebrate pests unit for supplying fox faeces, and to Brad Walker, Chad Staples and Featherdale

Wildlife Park for supplying quoll faeces. The research was carried out under the auspices of a NSW National Parks and Wildlife Service (NPWS) scientific research licence B2185 and UNSW Animal Care and Ethics Approval No. 00/64. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 165

6.7 References

Alexander, D. (1980). Diet of the Tiger Quoll (Dasyurus maculatus Kerr) in an area of New England National Park, Northeastern New South Wales. Honours thesis, University of New England, Armidale.

Barreto, G. R. and Macdonald, D. W. (1999). The response of water voles, Arvicola terrestris, to the odours of predators. Animal Behaviour 57, 1107-1112.

Belcher, C. A. (1995). Diet of the tiger quoll (Dasyurus maculatus) in East Gippsland, Victoria. Wildlife Research 22, 341-357.

Brown, J. S. (1988). Patch use as an indicator of habitat preference, predator risk, and competition. Behavioural Ecology and Socio-biology 22, 37-47.

Calder, C. J. and Gorman, M. L. (1991). The effects of Red Fox Vulpes vulpes faecal odours on the feeding behaviour of Orkney Voles Microtus arvalis. Journal of Zoology (London) 224, 599-606.

Caro, T. (2005). Antipredator defenses in birds and mammals. University of Chicago Press, Chicago.

Chambers, L. K. and Dickman, C. R. (2002). Habitat selection of the long-nosed bandicoot, Perameles nasuta (Mammalia, Peramelidae), in a patchy urban environment. Austral Ecology 27, 334-342.

Claridge, A. W., McNee, A., Tanton, M. T. and Davey, S. M. (1991). Ecology of bandicoots in undisturbed forest adjacent to recently felled logging coupes: a case study from the Eden Woodchip Agreement Area. In Conservation of Australia's Forest Fauna. (edited by D. Lunney). Royal Zoological Society of New South Wales, Mosman.

Claridge, A. W., Mifsud, G., Dawson, J. and Saxon, M. J. (2004). Use of infrared digital cameras to investigate the behaviour of cryptic species. Wildlife Research 31, 645-650.

Coman, B. J. (1998). Fox Vulpes vulpes. In The Mammals of Australia. (edited by R. Strahan) pp. 698-699. New Holland Publishers, Sydney.

Curio, E. (1976). The Ethology of Predation. Springer-Verlag, Berlin.

Dawkins, R. and Krebs, J. R. (1979). Arms races between and within species. Proceedings of the Royal Society of London B 205, 489-511.

Dickman, C. R. (1992). Predation and Habitat Shift in the House Mouse, Mus domesticus. Ecology 73, 313-322.

Dickman, C. R. and Doncaster, C. P. (1984). Responses of Small Mammals to Red Fox (Vulpes vulpes) Odour. Journal of Zoology (London) 204, 521-531. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 166

Edgar, R. and Belcher, C. A. (1998). Spotted Tailed quoll Dasyurus maculatus. In The Mammals of Australia. (edited by R. Strahan). New Holland Publishers, Sydney.

Epple, G., Mason, J. R., Nolte, D. L. and Campbell, D. L. (1993). Effects of predator odors on feeding in the mountain beaver (Aplodontia rufa). Journal of Mammalogy 74, 715-722.

Gordon, G. (1974). Movements and Activity of the Shortnosed Bandicoot Isoodon macrourus Gould (Marsupialia). Mammalia 38, 405-431.

Gorman, M. L. (1984). The response of prey to Stoat (Mustela erminea) scent. Journal of Zoology (London) 202, 419-423.

Green, K. and Osbourne, W. S. (1981). The diet of foxes Vulpes vulpes (L), in relation to abundance of prey above the winter snowline in New South Wales. Australian Wildlife Research 87, 239-250.

Gresser, S. (1996). Anti-predator Behaviour of the Common Brushtail Possum (Trichosurus vulpecula) at Burrendong Dam, N.S.W. Honours Thesis, University of Sydney, Sydney: 106.

Herman, C. S. and Valone, T. J. (2000). The effect of mammalian predator scent on the foraging behaviour of Dipodomys merriami. Oikos 91, 139-145.

Hornsby, P. E. (1984). Fauna Survey of Douglas Scrub. South Australian Naturalist 59, 14-17.

Jedrzejewski, W., Rychlik, W. and Jedrzejewska, B. (1993). Responses of bank voles to odours of seven species of predators: experimental data and their relevance to natural predator-vole relationships. Oikos 68, 251-257.

Jones, M. and Dayan, T. (2000). Foraging behaviour and microhabitat use by spiny mice Acomys cahirinus and A. russatus, in the presence of Blandford's fox (Vulpes vulpes) odor. Journal of Chemical Ecology 26, 455-469.

Jones, M. E. and Barmuta, L. A. (1998). Diet overlap and abundance of sympatric dasyurid carnivores: a hypothesis of competition? Journal of Animal Ecology 81, 434-447.

Kats, L. B. and Dill, L. M. (1998). The scent of death: Chemosensory assessment of predation risk by prey animals. Ecoscience 5, 361-394.

Kinnear, J. E., Sumner, N. R. and Olnus, M. L. (2002). The red fox in Australia-an exotic predator turned biocontrol agent. Biological Conservation 108, 335-359.

Kruuk, H. and Jarman, P. J. (1995). Latrine use by the Spotted-tailed quoll (Dasyurus maculatus: Dasyuridae, Marsupialia) in its natural habitat. Journal of Zoology (London) 236, 345-349.

Lima, S. L. and Dill, L. M. (1990). Behavioral decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology 68, 619-640. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 167

Lunney, D., Triggs, B., Eby, P. and Ashby, E. (1990). Analysis of scats of dogs Canis familiaris and foxes Vulpes vulpes (Canidae: Carnivora) in coastal forests near Bega, New South Wales. Australian Wildlife Research 17, 61-68.

Lyne, A. G. (1990). A brief review of bandicoot studies. In Bandicoots and Bilbies. (edited by J. H. Seebeck, P. R. Brown, R. L. Wallis and C. M. Kemper) pp. xxiii-xxix. Surrey Beatty and Sons, Sydney.

Macdonald, D. W. (1979). Some Observations and Field Experiments on the Urine Marking Behaviour of the Red Fox Vulpes vulpes L. Zeitshrift fuer Tierpsychologie 51, 1-22.

Macdonald, D. W. (1980). Patterns of Scent Marking with Urine and Faeces Amongst Carnivores Communities. Symposium of the Zoological Society of London 45, 107-139.

Maxwell, S., Burbidge, A. A. and Morris, K. (1996). The 1996 Action Plan for Australian Marsupials and Monotremes. Australasian Marsupial and Monotreme Specialist Group, IUCN Species Survival Commission, Canberra.

McKay, G. M. (1994). Effects of introduced predators on biodiversity remnant vegetation. Australian Nature Conservation Agency, Canberra.

Minitab (2003). Minitab Release 14.1, Statistical Software. Pennsylvania, Minitab.

Mitchell, B. D. and Banks, P. B. (2005). Do wild dogs exclude foxes? Evidence for competition from dietary and spatial overlaps. Austral Ecology 30, 581-591.

Noldus (2002). Ethovision 2.3, Video Tracking system for automation of behavioural experiments. Wageningen, Noldus Information Technology.

Nolte, D. L., Mason, J. R., Epple, G., Arnov, E. and Campbell, D. L. (1994). Why are predator urines aversive to prey. Journal of Chemical Ecology 20, 1505-1516.

Orrock, J. L., Danielson, B. J. and Brinkerhoff, R. J. (2004). Rodent foraging is affected by indirect, but not direct, cues of predation risk. Behavioural Ecology 15, 433- 437.

Parsons, G. J. and Bondrup-Nielsen, S. (1996). Experimental analysis of behaviour of meadow voles (Microtus pennsylvanicus) to odours of the short-tailed weasel (Mustela erminea). Ecoscience 3, 63-69.

Paull, D. (1995). The Distribution of the Southern Brown Bandicoot (Isoodon obesulus obesulus) in South Australia. Wildlife Research 22, 585-600.

Powell, F. and Banks, P. B. (2004). Do house mice modify their foraging behaviour in response to predator odours and habitat? Animal Behaviour 67, 753-759.

Preiss, K. and Kraehenbuehl, D. (1966). The Manning Fauna and Flora Reserve. South Australian Naturalist 41, 5-19. Chapter 6: Anti-predator strategies of bandicoots in response to predator odours 168

Richards, J. D. and Short, J. (2003). Reintroduction and establishment of the western barred bandicoot Perameles bougainville (Marsupialia: Peramelidae) at Shark Bay, Western Australia. Biological Conservation 109, 181-195.

Rolls, E. C. (1969). They All Ran Wild, the Animals and Plants that Plague Australia. Angus and Robertson, Sydney.

Seebeck, J. H. (1979). Status of the Barred Bandicoot, Perameles gunnii, in Victoria: with a Note on Husbandry of a Captive Colony. Australian Wildlife Research 6, 255-264.

Seebeck, J. H. (1990). Status, Distribution and Biogeography of the Eastern Barred Bandicoot, Perameles gunnii, in Victoria. In Management and Conservation of Small Populations. (edited by T. W. Clark and J. H. Seebeck) pp. 21-32. Chicago Zoological Society, Brookfield.

Short, J., Kinnear, J. E. and Robley, A. (2002). Surplus killing by introduced predators in Australia-evidence for ineffective anti-predator adaptations in native prey species? Biological Conservation 103, 283-301.

Southgate, R. I. (1990). Distribution and abundance of the Greater Bilby Macrotis lagotis Reid (Marsupialia: Peramelidae). In Bandicoots and Bilbies. (edited by J. H. Seebeck, P. R. Brown, R. L. Wallis and C. M. Kemper) pp. 293-302. Surrey Beatty and Sons, Sydney.

Stodart, E. (1966). Management and Behaviour of breeding groups of the marsupial Perameles nasuta Geoffroy in captivity. Australian Journal of Zoology 14, 611- 623.

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

Sullivan, T. P., Nordstrom, L. O. and Sullivan, D. S. (1985). Use of predator odors as repellents to reduce feeding damage by herbivores I. Snowshoe Hares (Lepus americanus). Journal of Chemical Ecology 11, 903-919.

Swihart, R. K. (1991). Modifying scent-marking behavior to reduce woodchuck damage to fruit trees. Ecological Applications 1, 98-103.

Thorsen, J. M., Morgan, R. A., Brown, J. S. and Norman, J. E. (1998). Direct and indirect cues of predatory risk and patch use by fox squirrels and thirteen-lined ground squirrels. Behavioural Ecology 9, 151-157.

Triggs, B. (1996). Tracks, scats and other traces; A field guide to Australian mammals. Oxford University Press, South Melbourne.

Ward, J. F., Macdonald, D. W. and Doncaster, C. P. (1997). Responses of foraging hedgehogs to badger odour. Animal Behaviour 53, 709-720.

Watts, C. H. S. (1969). Distribution and Habits of the Rabbit Bandicoot. Transactions of the Royal Society of South Australia 93, 135-141. Chapter 7: Volatiles of fox and quoll odour 169

Chapter 7:

Comparisons between the volatile constituents of red fox

Vulpes vulpes and tiger quoll Dasyurus maculatus urinary and faecal odours

7.1 Abstract

It has been suggested that prey species may respond to odour components common to many species of carnivore. To determine if the similar responses of native Australian prey species to the odours of the native tiger quoll Dasyurus maculatus and the introduced red fox Vulpes vulpes were due to such common components, I used Solid

Phase Microextraction (SPME) to compare the volatile constituents of the urinary and faecal odours of these predators. The odour profiles of the urine and faeces of the two species were significantly different. The only common components have also been found in a wide range of species, both predators and prey. Whilst prey species may respond to odour components common to closely related predator species, tiger quolls and red foxes are so evolutionarily distinct that their odours do not share any such common components.

7.2 Introduction

In the “attack - defence arms race” between predator and prey (Dawkins and Krebs

1979), many mammalian prey species have evolved to recognise olfactory cues from potential predators as an indication of increased predation risk and respond by modifying their behaviour to reduce this risk of predation (Kats and Dill 1998). The avoidance by Orkney voles Microtus arvalis of traps scented with stoat Mustela Chapter 7: Volatiles of fox and quoll odour 170

erminea anal gland scent (Gorman 1984) and red fox Vulpes vulpes faeces (Calder and

Gorman 1991) suggested that this response may be innate, as Orkney voles have been isolated from these predators for at least 4000 years. However, house mice Mus domesticus on predator-free islands in Western Australia did not avoid traps scented with domestic cat Felis catus and red fox faeces, after less than 100 years of isolation from these predators, although house mice on nearby islands where only cats were present avoided the odours of both predators (Dickman 1992). Dickman (1992) suggested this shared response was due to common components in the faecal odour of the two predators. Borowsksi (1998) similarly suggested that the response of Orkney voles to stoat odour might have been due to common components in the odour of stoats and other carnivores which are found in the Orkneys. The only wild mammalian predator in the Orkneys is the otter Lutra lutra (Calder and Gorman 1991), however a large number of Orkney voles are taken by domestic cats (Corbet and Wallis 1977).

Further evidence for a generalised response to common components of carnivore odour is inferred from the study of prey responses to novel predators. For example, field voles

M. agrestis avoid tiger Panthera tigris urine (Stoddart 1982a; Stoddart 1982b), water voles Arvicola terrestris avoid American mink M. vison urine and faeces (Barreto and

Macdonald 1999), red deer Cervus elaphus reduce their feeding in the presence of Lion

Panthera leo faeces (Abbott et al. 1990), and black-tailed deer eat less in the presence of lions, tigers and snow leopards Panthera unica (Muller-Scwarze 1972). The differing responses of bank voles to six species of carnivore indicate that they recognised each species’ distinctive odour (Jedrzejewski et al. 1993); however this does not preclude the possibility of an underlying common carnivore odour. Chapter 7: Volatiles of fox and quoll odour 171

In previous chapters, I revealed complex responses of small mammal prey to the faecal odours of a native marsupial predator, the tiger quoll Dasyurus maculatus and an introduced placental predator, the red fox. In Chapters 4 and 5 I showed that native rodent species responded equally to both tiger quoll and red fox faecal odours, even though foxes have been in Australia for less than 150 years (Rolls 1969). These responses support the common constituents hypothesis, but may also have occurred because the rodents possess some ability to recognise placental predators, as their ancestors evolved in Asia with placental predation pressure (Watts and Aslin 1981;

Strahan 1998). In Chapters 5 and 6, I showed that brown antechinus Antechinus stuartii and bandicoots respectively, also responded to red fox odour, but the responses were much weaker than to tiger quoll odour. These weaker responses to the novel predator also support the common constituents hypothesis, although predation pressure by foxes on bandicoots and antechinus has been substantial since their introduction 130 years ago (Rolls 1969), and may have provided enough selection pressure to induce mild avoidance.

Nolte et al. (1994) suggested that the repellency of carnivore odours may be due to sulfurous compounds derived from the digestion of meat, and found that the removal of sulfurous compounds from coyote urine Canis latrans significantly decreased its repellency to mountain beavers Aplodontia rufa. The volatile component from the anal gland scent marks of Mustela species are dominated by different combinations of a series of such sulfurous compounds (Sokolov et al. 1980; Brinck et al. 1983); some components have been found to be aversive to prey species, while others are not

(Sullivan and Crump 1984). Of nine compounds identified in the volatile constituents of red fox urine, including three sulfur containing compounds (Jorgenson et al. 1978; Chapter 7: Volatiles of fox and quoll odour 172

Wilson et al. 1978; Bailey et al. 1980), only 3-methyl-3-butenyl methyl sulfide was found to be aversive to snowshoe hares Lepus americanus, and to the same degree as actual red fox urine (Sullivan and Crump 1986). Sullivan and Crump (1986) suggested that the response of snowshoe hares and black-tailed deer Odocoileus hemionus columbianus to wolf Canis lupus urine may been because it also contains 3-methyl-3- butenyl methyl sulfide (Raymer et al. 1984; Raymer et al. 1986). This compound has also been found in the urine of coyotes (Schultz et al. 1988) and domestic dogs (Schultz et al. 1985), as well as the anal sac secretions of American mink (Sokolov et al. 1980) and has been reported to be in the volatile constituents of red fox faeces (Vernet-Maury et al. 1984). This suggests that it may be acting as a common component in the odours of a number of carnivore species. However, all these carnivores share a common lineage as members of the order Carnivora, as opposed to the tiger quoll, which is a member of the marsupial order Dasyuromorphia.

In this chapter, Solid Phase Microextraction (SPME) is used to analyse the volatile constituents of the urine and faeces of the red fox and the tiger quoll. SPME is a relatively new technique (Arthur and Pawliszyn 1990) which has simplified the sampling and analysis of headspace volatiles (Zhang and Pawliszyn 1993; Mills and

Walker 2000) and has previously been used by Toftegaards et al. (1999) to analyse the urinary volatiles of the brown antechinus. My specific aims are to:

1. analyse red fox urine to determine if the odour components determined by

Jorgenson et al. (1978) are detectable using this method;

2. analyse tiger quoll urine to determine if it contains 3-methyl-3-butenyl methyl

sulfide, or any other components in common with fox urine; and Chapter 7: Volatiles of fox and quoll odour 173

3. analyse red fox and tiger quoll faeces from the captive populations used in

Chapter 3-6 to determine if there are any common odour components which

native species may have been responding to.

7.3 Methods

Six samples of urine and faeces were analysed for each species. Tiger quoll urine was collected in plastic (polystyrene) 120 mL specimen jars from captive animals at

Featherdale Wildlife Park in Sydney’s west, when animals were being moved between enclosures. Tiger quolls habitually urinate when handled by humans. Faeces were collected from the cages. As mentioned in Chapters 3 and 4 these animals are fed a varied diet similar as to what they would have in the wild, including poultry, rat, rabbit, and macropod. Urine samples were taken from five males and one female; faeces were collected from three males and three females. Red fox faeces were collected from three male and three female captive animals at the CSIRO Sustainable Ecosystems Unit in

Canberra, fed on a diet of sheep and kangaroo carcasses and dog food. Urine was collected from the bladders of animals shot in western New South Wales near Orange, as sufficient urine for analysis was too difficult to obtain from the captive foxes. Three of the urine samples were from male foxes and three were from female foxes. All samples were collected and kept in sealed containers, frozen within one hour of collection and used within one month of collection.

Solid-phase microextraction (SPME) was used to sample the headspace of the samples.

For urine samples, a 0.5 mL aliquot was diluted with an equal volume of saturated sodium chloride solution in a 4 mL headspace vial (Toftegaards et al. 1999). For faecal samples 3g (±0.05) of faeces was placed in a 10 mL headspace vial. Samples were left Chapter 7: Volatiles of fox and quoll odour 174

to equilibrate for 20 minutes. A 65 μm polydimethylsiloxane/divinylbenzene fiber

(Supelco, Bellefonte, Pensylvania) was inserted through the vial septum and exposed to the headspace of the sample for 20 minutes. The SPME device was then inserted directly in the GC injector, and exposed as data acquisition began.

Chromatographic analysis was performed using an HP5890II GC using a DB-5MS column (30 m, 0.25 mm, 0.25 μm). Injection was splitless for one min, with the injector temperature set at 250°C. A 0.75 mm ID injection port liner (Supelco) was used to increase linear velocity, resulting in sharper peaks in the GC analysis. The fiber was left in the injector port for 10 minutes to allow for complete reconditioning. The column temperature was held at 40°C for 2 minutes, then raised to 260°C at 10°C/min and held at this temperature for 6 minutes (Toftegaards et al. 1999). Mass Spectra were acquired via a HP5971A Mass Selective Detector (MSD) from m/z 35 to m/z 550.

Volatile constituents of urine and faeces from foxes and quolls were compared using non-metric multidimensional scaling (MDS) and ANOSIM in Primer 5 (Primer-E

2001). It was beyond the scope of this study to undertake a complete analysis of the exact chemical formula of each compound. Therefore, the timing and height of each peak were used to characterise the sample (following Hayes et al. 2002). Data for each peak was untransformed in the generation of a Bray-Curtis similarity matrix for the

MDS plot.

7.4 Results

The timing of the major peaks was consistent within all six samples of urine or faeces for both red foxes and tiger quolls, although the relative amounts varied. While the Chapter 7: Volatiles of fox and quoll odour 175

peaks were similar within the same species, there was little crossover between the two species in terms of volatiles which were common to both species. Figures 7.1 to 7.4 are example chromatograms with the consistent peaks numbered.

Seven of the eight volatile components identified by Jorgenson et al. (1978) in fox urine were found in at least one of the six fox urine samples here, although only 4- heptanone, 3-methyl-3-butenyl methyl sulfide, Benzaldehyde, and Acetophenone

(peaks 1 ,4, 5 and 10 respectively) were found in all six. The eighth volatile component

2-methyl quinoline was however identified in fox faecal samples (peak 15), although as for Jorgenson et al. (1978) only in male animals. 3-methyl-3-butenyl methyl sulfide was also found in high concentrations in fox faeces (peak 7), but was absent from the urine and faeces of the tiger quolls.

There was little similarity between the urine and faeces of red foxes and tiger quolls;

MDS analysis showed that the mixture of volatiles from each sample was more similar within each group (fox urine, quoll urine, fox faeces, quoll faeces) than any of the samples were to samples from other groups (ANOSIM: Global R = 1, P < 0.002)

(Figure 7.5). The one outlying sample for quoll faeces contained fewer of the smaller peaks, but still contained the same major peaks as the other quoll faecal samples, and the Global R of 1 indicates that it was still more similar to the other quoll faecal samples than any of the other groups.

The few common components in red fox and tiger quoll urine and faeces are summarised in Table 7.1. For urine, based on comparisons with known mass spectra, the common components were benzaldehyde and acetophenone. For faeces, the Chapter 7: Volatiles of fox and quoll odour 176

common components were indole and butanoic acid, 3-methyl butanoic acid, and 2- methyl butanoic acid, although the latter three compounds were only found in the faeces of female tiger quolls.

Table 7.1: Common volatile components of fox and quoll faeces and urine, and their peaks as numbered in figs 7.1-7.4 Chemical Fox Peak Quoll Peak Urine Benazaldehyde 5 7 Acetophenone 10 10 Faeces Butanoic Acid 3 3 3-methyl butanoic Acid 4 6 2-methyl butanoic Acid 5 7 Indole 14 23

Figure 7.1: Chromatogram of the headspace volatile compounds from red fox urine. Major peaks are numbered 1-17. Chapter 7: Volatiles of fox and quoll odour 177

Figure 7.2: Chromatogram of the headspace volatiles compounds from tiger quoll urine. Major peaks are numbered 1-23.

Figure 7.3: Chromatogram of the headspace volatile compounds from red fox faeces. Major peaks are numbered 1-17. Chapter 7: Volatiles of fox and quoll odour 178

Figure 7.4: Chromatogram of the headspace volatile compounds from tiger quoll faeces. Major peaks are number 1-24.

Figure 7.5: Two dimensional MDS ordination of odour components of red fox and tiger quoll urine and faeces, based on a square root transformation and a Bray-Curtis similarity matrix. All 24 samples are shown, six for each group, however for the same group most are clustered at exactly the same point. Chapter 7: Volatiles of fox and quoll odour 179

7.5 Discussion

The results indicate that SPME is an appropriate method to analyse the volatile constituents of mammalian urinary and faecal odours. The components detected by previous more expensive and labour intensive methods requiring greater volumes of sample were also detected using SPME (Jorgenson et al. 1978).

As expected, 3-methyl-3-butenyl methyl sulfide was detected in all samples of red fox urine and faeces, but it was not detected in any of the tiger quoll samples. This compound has been proposed as a key common constituent causing avoidance in a range of prey species to novel predators (Sullivan et al. 1985). However, my results show that the similar responses of the native Australian species to red fox and tiger quoll faeces in Chapters 4, 5 and 6 were not due to this particular odour component.

Indeed, the mixture of volatile components was significantly different between red fox and tiger quoll faeces, and the few common components also occur in many other substances of non-predator origin, and hence are unlikely to be a useful cue to predation. For example, Arnould (1998) found that indole was not responsible for the repellency of domestic dog Canis lupus familiaris faeces, and suggested that this was because it is observed in numerous biological products from many different species and origins (Albone 1984), including corn Zea mays (Turlings et al. 1998; Gouinguene and

Turlings 2002) and the milk of dairy cows Bos taurus (Friedrich and Acree 1998), as well as the faeces of humans Homo sapiens (Moore et al. 1987), domestic pigs Sus scrofa (Yasuhara et al. 1984) and lions P. leo (Abbott et al. 1990). The butanoic acid compounds were detected for both predators, but only at very low concentrations, and have been reported in the volatile component of house mice (Goodrich et al. 1990), Chapter 7: Volatiles of fox and quoll odour 180

European rabbit Oryctolagus cuniculus (Goodrich et al. 1981), and domestic pig faeces

(Yasuhara et al. 1984) as well as lion (Abbott et al. 1990) and dog faeces (Arnould et al. 1998). Likewise the common components of red fox and tiger quoll urine, acetophenone and benzaldehyde, have been found in the urine of a range of both predator and prey species (Matsumoto et al. 1973; Raymer et al. 1984; Raymer et al.

1986; Schwende et al. 1986; Schultz et al. 1988; Boyer et al. 1989; Andersen and

Vulpius 1999; Ma et al. 1999; Toftegaards et al. 1999).

Although 3-methyl-3-butenyl methyl sulfide is not responsible for the reaction of

Australian prey species to tiger quoll odour, it may still be acting as a common chemical “trigger” (Dickman and Doncaster 1984) among placental carnivores. It is known to be present in the urine of foxes, wolves (Raymer et al. 1984; Raymer et al.

1986), coyotes (Schultz et al. 1988), and domestic dogs (Schultz et al. 1985), the anal sac secretions of American mink (Sokolov et al. 1980) and fox faeces, and may be present in the urine and faeces of a number of other carnivore species as well. It was not found in the urine of either the bobcat Lynx rufus (Mattina et al. 1991) or the lion

(Andersen and Vulpius 1999), suggesting it is not present in the urine of felids, and is not ubiquitous throughout the Carnivora, however prey species may also respond to odour components common within the Felidae. This would explain why field voles respond to tiger urine (Stoddart 1982a; Stoddart 1982b), as they are commonly preyed upon by wildcats F. silvestris (Hewson 1983).

Native rodents react very similarly to the faecal odour of red foxes and tiger quolls

(Chapter 4 and 5), and antechinus and bandicoots also responded to the odours of both predators (Chapters 5 and 6). The chemical analysis here suggests that these responses Chapter 7: Volatiles of fox and quoll odour 181

are not due to common components in these predator odours. Nor are these prey species likely to be responding to odour compounds they are unfamiliar with simply because they are sulfur-based, as prey species respond to certain sulfur-based chemicals and not others, even when they are similar in structure and/or are isolated from the same odour source (Sullivan and Crump 1984; Vernet-Maury et al. 1984; Sullivan and Crump

1986). These results reject the common constituents hypothesis, suggesting more complex causes of predator avoidance in these Australian native mammals. Sulfurous or otherwise, the structure of the volatile components of tiger quoll odour to which prey species are responding have yet to be identified as a comprehensive structural analysis was beyond the scope and resources of this particular study. However, it would be interesting to determine if there are common odour components among related marsupial carnivores to which prey species are responding, as there appear to be for placental predators.

7.6 Acknowledgements

I would like to thank Chad Staples from Featherdale Wildlife Park for supplying tiger quoll urine and faeces, and Steve Henry from the CSIRO Sustainable Ecosystems unit for supplying fox faeces. A big thank you to Roy Bladen for supplying fox urine, after I encountered several other supply problems. Thanks also to Rocky de Nys and Tim

Charlton for technical advice. And an enormous thank you to Neda Shakibaee for access to equipment, technical advice and assistance in performing the GCMS analysis.

Greatly appreciated. Chapter 7: Volatiles of fox and quoll odour 182

7.7 References

Abbott, D. H., Baines, D. A., Faulkes, C. G., Jennings, D. C., Ning, P. C. Y. K. and Tomlinson, A. J. (1990). A natural deer repellent: chemistry and behaviour. In Chemical signals in vertebrates 5. (edited by D. W. Macdonald, D. Muller- Schwarze and S. E. Natynczuk). Oxford University Press, Oxford.

Albone, E. S. (1984). Mammalian Semiochemistry, the investigation of chemical signals between mammals. John Wiley and Sons Ltd, Chichester.

Andersen, K. F. and Vulpius, T. (1999). Urinary Volatile Constituents of the Lion Panthera leo. Chemical Senses 24, 179-189.

Arnould, C., Malosse, C., Signoret, J.-P. and Descoins, C. (1998). Which chemical constituents from dog feces are involved in its food repellent effect in sheep? Journal of Chemical Ecology 24, 550-576.

Arthur, C. L. and Pawliszyn, J. (1990). Solid Phase Microextraction with Thermal Desorption Using Fused Silica Optical Fibers. Analytical Chemistry 62, 2145- 2148.

Bailey, S., Bunyan, P. J. and Page, J. M. J. (1980). Variation in the levels of some components of the volatile fraction of urine from captive red foxes (Vulpes vulpes) and its relationship to the state of the animal. In Chemical Signals: Vertebrates and Aquatic Invertebrates. (edited by D. Muller-Schwarze and R. M. Silverstein). Plenum Press, New York.

Barreto, G. R. and Macdonald, D. W. (1999). The response of water voles, Arvicola terrestris, to the odours of predators. Animal Behaviour 57, 1107-1112.

Borowski, Z. (1998). Influence of predator odour on the feeding behaviour of the root vole (Microtus oeconomus Pallas 1776). Canadian Journal of Zoology 76, 1791-1794.

Boyer, M. L., Jemiolo, B., Andreolini, F., Weisler, D. and Novotny, M. (1989). Urinary volatile profiles of Pine Vole Microtus pinetorum, and their endocrine dependency. Journal of Chemical Ecology 15, 649-662.

Brinck, C., Erlinge, S. and Sandell, M. (1983). Anal sac secretion in Mustelids: A comparison. Journal of Chemical Ecology 9, 727-745.

Calder, C. J. and Gorman, M. L. (1991). The effects of Red Fox Vulpes vulpes faecal odours on the feeding behaviour of Orkney Voles Microtus arvalis. Journal of Zoology (London) 224, 599-606.

Corbet, G. B. and Wallis, R. L. (1977). Orkney/Guernsey vole Microtus arvalis. In The Handbook of British Mammals. (edited by G. B. Corbet and H. N. Southern). Blackwell Scientific Publications, Oxford.

Dawkins, R. and Krebs, J. R. (1979). Arms races between and within species. Proceedings of the Royal Society of London B 205, 489-511. Chapter 7: Volatiles of fox and quoll odour 183

Dickman, C. R. (1992). Predation and Habitat Shift in the House Mouse, Mus domesticus. Ecology 73, 313-322.

Dickman, C. R. and Doncaster, C. P. (1984). Responses of Small Mammals to Red Fox (Vulpes vulpes) Odour. Journal of Zoology (London) 204, 521-531.

Friedrich, J. E. and Acree, T. E. (1998). Gas Chromatography Olfactometry (GC/O) of Dairy Products. International Dairy Journal 8, 235-241.

Goodrich, B. S., Gambale, S., Pennycuik, P. R. and Redhead, T. D. (1990). Volatiles from feces of wild male house mice. Journal of Chemical Ecology 16, 2091- 2106.

Goodrich, B. S., Hesterman, E. R., Shaw, K. S. and Mykytowycz, R. (1981). Identification of some volatile compounds in the odor of fecal pellets of the rabbit Oryctolagus cuniculus. Journal of Chemical Ecology 7, 817-827.

Gorman, M. L. (1984). The response of prey to Stoat (Mustela erminea) scent. Journal of Zoology (London) 202, 419-423.

Gouinguene, S. P. and Turlings, T. C. J. (2002). The Effects of Abiotic Factors on Induced Volatile Emissions in Corn Plants. Plant Physiology 129, 1296-1307.

Hayes, R. A., Richardson, B. J. and Wyllie, S. G. (2002). Semiochemicals and social signalling in the wild European rabbit in Australia: I. Scent profiles of chin gland secretion from the field. Journal of Chemical Ecology 28, 363-384.

Hewson, R. (1983). The food of wild cats (Felis silvestris) and red foxes (Vulpes vulpes) in west and north-east Scotland. Journal of Zoology (London) 200, 283- 289.

Jedrzejewski, W., Rychlik, W. and Jedrzejewska, B. (1993). Responses of bank voles to odours of seven species of predators: experimental data and their relevance to natural predator-vole relationships. Oikos 68, 251-257.

Jorgenson, J. W., Novotny, M., Carmack, M., Copeland, G. B., Wilson, S. R., Katona, S. and Whitten, W. K. (1978). Chemical scent constituents in the urine of the red fox (Vulpes vulpes) during the winter season. Science 199, 796-798.

Kats, L. B. and Dill, L. M. (1998). The scent of death: Chemosensory assessment of predation risk by prey animals. Ecoscience 5, 361-394.

Ma, W., Weisler, D. and Novotny, M. V. (1999). Urinary volatile profiles of the deermouse (Peromyscus maniculatus) pertaining to gender and age. Journal of Chemical Ecology 25, 417-431.

Matsumoto, K. E., Partridge, D. H., Robinson, A. B. and Pauling, L. (1973). The identification of volatile compounds in human urine. Journal of Chromatography 85, 31-34. Chapter 7: Volatiles of fox and quoll odour 184

Mattina, M. J. I., Pignatello, J. J. and Swihart, R. K. (1991). Identification of volatile components of Bobcat (Lynx rufus) urine. Journal of Chemical Ecology 17, 451-462.

Mills, G. A. and Walker, V. (2000). Headspace solid-phase microextraction procedures for gas chromatographic analysis of biological fluids and materials. Journal of Chromatography A 902, 267-287.

Moore, J. G., Jessop, L. D. and Osbourne, D. N. (1987). Gas-Chromatographic and Mass-Spectrometric Analysis of the odour of Human Feces. Gastroenterology 93, 1321-1329.

Muller-Scwarze, D. (1972). Responses of young black-tailed deer to predator odors. Journal of Mammalogy 53, 393-394.

Nolte, D. L., Mason, J. R., Epple, G., Arnov, E. and Campbell, D. L. (1994). Why are predator urines aversive to prey. Journal of Chemical Ecology 20, 1505-1516.

Primer-E (2001). Primer 5 for Windows, Version 5.2.2. Plymouth, Primer-E.

Raymer, J., Wiesler, D., Novotny, M., Asa, C., Seal, U. S. and Mech, L. D. (1984). Volatile constituents of wolf (Canis lupus) urine as related to gender and season. Experientia 40, 707-709.

Raymer, J., Wiesler, D., Novotny, M., Asa, C., Seal, U. S. and Mech, L. D. (1986). Chemical scent constituents in urine of wolf (Canis lupus) and their dependence on reproductive hormones. Journal of Chemical Ecology 12, 297-314.

Rolls, E. C. (1969). They All Ran Wild, the Animals and Plants that Plague Australia. Angus and Robertson, Sydney.

Schultz, T. H., Flath, R. A., Stern, D. J., Mon, T. R., Teranishi, R., Kruse, S. M., Butler, B. and Howard, W. E. (1988). Coyote Estrous Urine Volatiles. Journal of Chemical Ecology 14, 701-712.

Schultz, T. H., Kruse, S. M. and Flath, R. A. (1985). Some volatile components of female dog urine. Journal of Chemical Ecology 11, 169-175.

Schwende, F. J., Weisler, D., Jorgenson, J. W., Carmack, M. and Novotny, M. (1986). Urinary volatile constituents of the house mice, Mus musculus, and their endocrine dependency. Journal of Chemical Ecology 12, 277-296.

Sokolov, V. E., Albone, E. S., Flood, P. F., Heap, P. F., Kagan, M. Z., Vasilieva, V. S., Roznov, V. V. and Zinkevich, E. P. (1980). Secretion and secretory tissues of the anal sac of the mink, Mustela vison. Journal of Chemical Ecology 6, 805- 825.

Stoddart, D. M. (1982a). Demonstration of Olfactory Discrimination by the Short-tailed Vole, Microtus agrestis L. Animal Behaviour 30, 293-301.

Stoddart, D. M. (1982b). Does trap odour influence estimation of population size of the short-tailed vole, Microtus agrestis? Journal of Animal Ecology 51, 375-386. Chapter 7: Volatiles of fox and quoll odour 185

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

Sullivan, T. P. and Crump, D. R. (1984). Influence of mustelid scent-gland compounds on suppression of feeding by snowshoe hares (Lepus americanus). Journal of Chemical Ecology 10, 1809-1821.

Sullivan, T. P. and Crump, D. R. (1986). Feeding responses of snowshoe hares (Lepus americanus) to volatile constituents of red fox (Vulpes vulpes) urine. Journal of Chemical Ecology 12, 729-739.

Sullivan, T. P., Nordstrom, L. O. and Sullivan, D. S. (1985). Use of predator odors as repellents to reduce feeding damage by herbivores II. Black-Tailed Deer (Odocoileus hemionus columbianus). Journal of Chemical Ecology 11, 921-935.

Toftegaards, C. L., Moore, C. and Bradley, A. J. (1999). Chemical Characterisation of Urinary Pheromones in Brown Antechinus. Journal of Chemical Ecology 25, 527-534.

Turlings, T. C. J., Lengwiler, U. B., Bernasconi, M. L. and Wechsler, D. (1998). Timing f induced volatile emissions in maize seedlings. Planta 207, 146-152.

Vernet-Maury, E., Polak, E. H. and Demael, A. (1984). Structure-activity relationship of stress-inducing odorants in the rat. Journal of Chemical Ecology 10, 1007- 1018.

Watts, C. H. S. and Aslin, H. J. (1981). The Rodents of Australia. Angus and Robertson Publishers, Sydney.

Wilson, S. R., Carmack, M., Novotny, M., Jorgenson, J. W. and Whitten, W. K. (1978). delta3-Isopentenyl Methyl Sulfide. A New Terpenoid in the Scent Mark of the Red Fox (Vulpes vulpes). Journal of Organic Chemistry 43, 4675-4676.

Yasuhara, A., Fuwa, K. and Jimbu, M. (1984). Identification of Odorous Compounds in Fresh and Rotten Swine Manure. Agricultural and Biological Chemistry 48, 3001-3010.

Zhang, Z. and Pawliszyn, J. (1993). Headspace solid-phase microextraction. Journal of Chemical Ecology 65, 1843-1852. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 186

Chapter 8:

The role of odour in the prey searching behaviour of tiger quolls Dasyurus maculatus

8.1 Abstract

It is widely assumed that mammalian predators rely heavily upon their sense of smell for detection of prey, however this has largely remained an assumption, and the manner in which olfaction is utilised is poorly recognised. This chapter investigates how captive individuals of the Australian marsupial predator species, the tiger quoll

Dasyurus maculatus, responded to the odours of a familiar food item and the urinary and faecal odours of prey species with which they share an evolutionary heritage. When the odour emanated from a point source, the captive quolls only responded to the familiar food item when they were within approximately 30 cm of the odour source and showed no response to the urinary and faecal odours. When an odour trail was created leading back to the primary source the quolls rapidly followed it, and spent more time in the area where the food item or the urine and faeces were. For tiger quolls in still air, olfaction appears to be effective only when they are in close proximity to an odour source. This is in accordance with the few other species whose use of olfaction in predation has been investigated. Once located however, these odour sources likely reveal information to the predators which allow them to modify their behaviour to more effectively locate their prey, utilising their senses in an integrated and systematic search. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 187

8.2 Introduction

Our understanding of predator/prey interactions has greatly improved over the last few decades, particularly in terms of the often complex behavioural responses of prey species to predation risk (Lima and Dill 1990). However, the role of predators in these interactions has largely been ignored (Lima 2002). This bias is especially evident in research on the importance of odour cues in mammalian predator-prey interactions. In their review of prey species responses to predator odours, Kats and Dill (1998) identified 78 studies where mammalian prey species altered their behaviour in response to the odours of mammalian predators (see also chapter 1). In contrast, only a handful of studies have examined how mammalian predator species utilise odour in their search for prey.

Several lab based studies have demonstrated the importance of olfaction in the near detection of live prey. Olfaction was found to be more important than vision or audition in the location of invertebrate prey within a confined space by predatory small mammals such as shrews (Holling 1958), spiny mice Acomys cahirinus (Langley 1988), and dasyurids (Huang 1986). Similar experiments on coyotes Canis latrans searching for rabbits in small windless enclosures found vision to be the most important of the three senses, while olfaction was the least important (Wells and Lehner 1978): however, in larger enclosures open to air movement, olfaction was found to be far more important than hearing, and only slightly less important than vision (Wells 1978). In contrast, Österholm (1964) considered hearing to be the most important of these three senses in the red fox Vulpes vulpes, but recognised that all three were utilized concurrently, and that their relative importance was dependent on current conditions. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 188

Each of these studies examined the direct location of prey in a confined space and only considered predator behaviour during the final phases of the predator/prey interaction, once attack was imminent. However, the predator/prey interaction begins much earlier, when predators are searching for profitable feeding opportunities (Endler 1991), and during this time olfaction may play a more vital role in revealing otherwise cryptic prey. Mammalian species produce a range of unavoidable odorous wastes, and many species also use deliberate scent marks to communicate with conspecifics, mark food sources and identify territorial boundaries (Macdonald and Brown 1985). These odorous secretions have the potential to attract scent hunting predators, and may be associated with focal points of activity such as odour accumulations at nest sites or foraging areas (Banks et al. 2002). Alternatively, odour sources may be distributed along trails that are frequently used for movement and/or scent marking of territory boundaries (Johnson 1975; Rozenfeld et al. 1987). However, only two studies have experimentally investigated how mammalian predators may exploit the scent marks of potential prey species. Cushing (1984) showed that weasels Mustela nivalis prefer the urinary odour of oestrous over dioestrous prairie deermice Peromyscus maniculatus bairdi, and Ylönen et al. (2003) showed that weasels selectively investigate areas that have been scented with the urine of bank voles Clethrionomys glareolus and field voles

Microtus agrestis over areas which have not. In this chapter, I examine the role of odour in the prey searching behaviour of tiger quolls Dasyurus maculatus.

Green and Scarborough (1990) suggested that tiger quolls rely greatly on their olfactory system, as they “constantly sniff the ground and their surroundings”. They have also been reported to be able pick up the cross-trail of a rabbit (Fleay 1932), indicating that they respond to the odours of potential prey species. Understanding the role of olfactory Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 189

cues in the location of prey by tiger quolls may have important implications in terms of interpreting the responses of prey species to the predation risk presented by these predators. Moreover, non-target uptake of fox/dog baits by quolls is a major conservation issue in Australia (Belcher 1998; Glen and Dickman 2003; Kortner et al.

2003; Murray and Poore 2004), requiring a deeper understanding of how quolls detect food sources, in order to develop more quoll friendly baiting technique. This chapter uses a series of three experiments to determine how tiger quolls utilise odour in their predatory explorations by investigating how they respond to the odour of a familiar food item and the urinary and faecal odours of common prey species; specifically how quickly they approach these odour sources, and how long they spend at the odour source, compared to an odourless control.

8.3 Methods

The experiments were conducted on captive tiger quolls at Featherdale Wildlife Park in

Sydney’s west. The six male animals used in the experiments were kept singly, in enclosures the same size as the experimental enclosure. All six animals were born at

Featherdale, from parents originally caught in the wild. The quolls are fed daily with a varied diet similar as to what they would have in the wild, including poultry, rat, rabbit, and macropod. Trials took place prior to the animal’s being fed each day. Although considered typically nocturnal, wild tiger quolls are often also active during the daytime

(Fleay 1932; Green and Scarborough 1990; Edgar and Belcher 1998; Burnett 2000), as are the captive quolls at Featherdale (pers. obs). Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 190

8.3.1 Experimental enclosure

The experimental enclosure was 230 cm wide and 250 cm deep, with a sloped roof such that the back wall was 190 cm tall and the front wall was 200 cm tall. The back and side walls were fully enclosed, the front was open mesh, and the enclosure had an open earth floor. A wooden experimental apparatus was designed to measure the quolls odour choices. It consisted of two 120 cm tall 50 x 50 cm platforms, connected by a

130 cm long, 15 cm wide cross piece and was placed at the back of the enclosure.

Access to the apparatus was via a 30 cm wide 200 cm long ramp, sloping upwards from the ground to the centre of the crosspiece, such that after climbing to the top of the ramp the animals had an equal chance of choosing to investigate either the left or right platform. In the centre of each platform was a 20 x 20 x 10 cm wooden box, with a hinged lid. Wire mesh 10 x 10 cm in the lid, and 5 x 10 cm on either side of the box allowed for diffusion of the odour from the source placed inside, without allowing the quolls to see or access the odour source (Figure 8.1).

8.3.2 Trial procedure

In each trial, a single tiger quoll was released at the entrance to the experimental enclosure and videoed for 25 minutes, using a Hitachi Handycam mounted on a tripod outside the front of the cage. The videos were later reviewed to determine two measures of interest in the odour; the time taken to travel from the top of the ramp to each of the platforms, and the time spent at each of the platforms. Three experiments were conducted; each consisted of two series of six trials. Within each series of trials the odour source was placed at each platform an equal number of times, but the order in which the left and right platforms were used was determined randomly, as was the order in which the six tiger quolls were used. A different sample of the odour source Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 191

(a)

(b)

Figure 8.1: Experimental apparatus with tiger quoll (a) climbing central ramp, and (b) approaching platform. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 192

was used for each trial and between trials the apparatus was thoroughly washed to remove any residual odours. Each series of six trials used a different odour source and was conducted on a separate day, with at least one week in between.

The first experiment tested whether tiger quolls would respond to odours emanating from a point source. The two odour sources were: a food source with a strong odour, with which they were familiar; and the urinary and faecal odour of an evolutionarily familiar prey species. In the first series of trials, the odorous food item was a whiting

Sillago sp., a regular part of the diet of quolls at Featherdale; fish are relished by captive quolls (Fleay 1932) and are known to be consumed by quolls in the wild

(Debbie Andrew pers. com.). A single whiting was placed in the box as the odour source. The other box was left empty. In the second series of trials, the evolutionarily familiar prey species was the bush rat Rattus fuscipes, one of the most common prey items in the diet of wild quolls (Belcher 1995; Burnett 2000). Cotton wool used as bedding from Elliott traps which had caught bush rats in Chapter 4 and was visibly contaminated with urine and faeces was placed in the box as the odour source. Cotton wool from three separate traps was used to provide a concentrated odour source. An equivalent amount of clean cotton wool was placed in the other box, to control for the odour of cotton wool.

The second experiment examined whether tiger quolls showed interest in odour trails and whether this decreased their search time to find the odour source. The two odour sources were again whiting and cotton wool soiled with urine and faeces of bush rats.

The procedure was the same as in the first experiment, except that an odour trail was first created by dragging the fish or cotton wool along the crosspiece from the centre to Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 193

the box where the odour source was then placed. For the trials with the bush rat bedding, clean cotton wool was also dragged from the centre of the cross piece to the other box to again control for the odour of cotton wool.

The third experiment expanded the range of odours tested to include the urinary and faecal odour of another evolutionarily familiar prey species and a strong odour not associated with potential prey. The other evolutionarily familiar prey species was the brown antechinus Antechinus stuartii, also one of the most common prey species in the diet of wild tiger quolls (Alexander 1980; Belcher 1995). As for bush rats, the actual odour source was cotton wool from Elliott traps which had caught brown antechinus in

Chapter 4, and was visibly contaminated with urine and faeces. The non-prey based odour was commercial eucalyptus oil (Bosisto’s). Several drops of eucalyptus oil were allowed to soak into an equivalent amount of cotton wool as used in the other trials. To the human nose the strength of the odour was comparable to that of the soiled bedding.

The trials were conducted as in the second experiment; an odour trail was created by dragging the cotton wool along the crosspiece, and clean cotton wool was used as a control.

8.3.3 Data analysis

The data from each series of trials was analysed separately. The time taken in seconds to reach the treatment and control platforms from the middle of the crosspiece was compared using a paired T-test. Similarly, a paired T-test was used to compare the total time spent at each platform in seconds. Quolls showing interest in a particular odour would have a lower time to reach the platform with that odour source, and would spend more time at the platform with the odours source than the other platform. Before Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 194

analysis, the data was log-transformed to improve normality. Statistical analyses were conducted using Minitab 14 (Minitab 2003).

8.4 Results

The time taken for quolls to reach the odour platform did not differ from the time taken to reach the control platform, regardless of which of the two point source odours was used (fish; t5=0.71, p=0.517: bush rat; t4=1.08, p=0.359), and the choice of which platform the animals first inspected appeared to be random (Figure 8.2a). However, they did spend more time at the odour platform than the other platform when the odour source was fish (t5=4.17, p=0.014), but not when the odour source was bush rat bedding

(t4=0.42, p=0.703) (Figure 8.3a). For both odour sources, one animal of the six (a different animal for each odour source) climbed to the top of the ramp but did not proceed to either platform, and hence was removed from the analyses. When bush rat bedding was used, one animal climbed to the top of the ramp and proceeded to the non- odour platform, where he spent a total of 16 seconds. As this animal did not spend any time at the odour platform he was also removed from the analysis.

In the experiment where an odour trail was created using fish or bush rat bedding, quolls rapidly detected the trail at the top of the ramp and followed it to the odour source, taking less time on average to approach the odour platform than the other platform (fish; t6=2.73, p=0.041: bush rat; t6=3.24, p=0.023) (Figure 8.2b), and spent more time at the odour platform for both odour sources (fish; t6=3.17, p=0.025: bush rat; t6=2.73, p=0.041) (Figure 8.3b). In all 12 trials the tiger quolls visited both platforms and on all but one approached the odour platform first. The only exception occurred using a fish trail when the quoll leapt across from the middle of the ramp to Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 195

the other platform, but then walked back across the crosspiece, picking up the odour trail at the middle, and nosing his way along it back to the odour platform.

In the third experiment, when brown antechinus bedding was used as the odour source, all six tiger quolls followed the odour trail and approached the odour platform more quickly than the other platform (t6=13.96, p<0.001) (Figure 8.2c), and spent more time in total at the odour platform (t6=3.24, p=0.023) (Figure 8.3c). However, when eucalyptus oil was used as the odour source they did not approach the odour source quicker (t4=1.2, p=0.317) (Figure 8.2c), and spent the same amount of time on the two platforms (t4=0.31, p=0.775) (Figure 8.3c). Two animals did not approach either platform despite climbing to the top of the ramp, and were removed from the analyses.

Two animals investigated the other platform first, and two animals investigated the odour platform first, although they did not follow the odour trail. These two quolls jumped off the apparatus after investigating the odour platform, and did not climb the apparatus and investigate the other platform until towards the end of the trial, hence the very large mean and standard error for the difference in time taken to approach the two platforms (Figure 8.2c). Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 196

(a)

40 30 20 10 0 -10 time (seconds) -20 -30 whiting bush rat

(b)

0 -25 -50 -75 -100 -125 -150 time (seconds) -175 -200 whitin bush rat

(c)

0 0 -25 -200 -50 -75 -400 -100 -125 -600

time (seconds) -150 -800 -175 -200 brown antechinus -1000 eucalyptus oil

Figure 8.2: The mean difference ± SE between the time (in seconds) taken by the tiger quolls to reach the odour platform from the top of the ramp, and the time taken to reach the other platform from the top of the ramp when there were (a) no trails, (b) odour trails added, (c) odour trails added. Negative values indicate less time taken to reach the odour platform. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 197

(a)

350 300 250 200 150 100 50 time (seconds) 0 -50 whiting bush rat

(b)

140 120 100 80 60 40

time (seconds) time 20 0 whiting bush rat

(c)

100

50

0

-50 time (seconds)time

-100 brown antechinus eucalyptus oil

Figure 8.3: The mean difference ± SE between the time (in seconds) spent by the tiger quolls on the odour platform and the time spent at the other platform when there were (a) no trails, (b) odour trails added, (c) odour trails added. Positive values indicate more time spent at the odour platform. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 198

8.5 Discussion

This is the first study to demonstrate that tiger quolls use odour in their foraging explorations. When odour trails were presented, quolls responded to both the odour of a familiar food item, and the urinary and faecal odours of evolutionary familiar prey species. They followed the trails to the odour-treated platform, investigating this platform first, and spending more time there than the other platform. They did not respond to the odour of eucalyptus oil. This indicates that quolls do not simply respond to every strong odour they encounter. Rather, they were specifically responding to the odour of the whiting and the urine and faeces of bush rats and brown antechinus.

The following of scent trails has been noted in a number of other carnivore species.

Spotted hyenas Crocuta crocuta (Bearder 1977) and brown hyenas Hyaena brunnea

(Mills 1978) follow the scent trails created by dragging a carcass. Wolves C. lupus

(Mech 1970) and American martens Martes americana (Spencer and Zielinski 1983) have been recorded following the sent trails of living prey, sniffing at their individual footprint. In each of these cases the predators were able to successfully follow the scent trail it until they found the potential food source, either the carcass or the living prey.

But in order to do so they first had to locate the odour trail.

Without the presence of the odour trail, the quolls could not detect the odour sources from the middle of the ramp, a distance of 65 cm, and did not appear to detect them until they were within approximately 30 cm. Such small detection distances have been noted previously in other mammalian predator species. Raber (1944) considered olfaction in the beech marten M. foina to only be effective in the near location of food.

Similarly, Österholm (1964) argued that the effective distance of the red fox’s sense of Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 199

smell was much less than is commonly assumed, describing it as only being important in the “point blank” location of prey. Although foxes sometimes located pieces of meat by olfaction up to a distance of 1.5 m, he stated that these pieces of meat also often remained unobserved.

It is possible that such small scale detection was due to the experimental set-up which would have restricted airflow though the cage. The distance over which predators can detect prey is dependent upon wind speed. For example, as long as they were downwind, coyotes were able to detect rabbits up to a distance of 6 m away in 50 km/h winds, but this distance of detection dropped to less than 1.5 m when wind speeds were below 10 km/h (Wells 1978). The detection of vole urine by weasels from a distance of

60 cm was also wind-assisted, as a fan was placed behind the odour sources (Cushing

1984; Ylönen et al. 2003). If the experiment had been conducted in an enclosure more open to the movement of air, the quolls may have been able to detect the odour sources over a greater distance. However, this would only have been the case if there was a wind blowing, and the quolls were downwind of the odour source.

Regardless of whether there was an odour trail or not, when whiting was used as the odour source tiger quolls spent more time at that platform than the other. Having located the fish by its odour, they attempted to gain access to the whiting by clawing at the mesh, and biting at the wood of the box. The reaction to bush rat and brown antechinus bedding was markedly different. Tiger quolls followed the odour trail back to the box, but rather than attempting to gain entry, they continued to sniff all over and around the box. Occasionally they would sit still for a few moments, but remaining vigilant, before resuming sniffing the platform. This is in stark contrast to how they Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 200

behaved towards the non-odour box and when eucalyptus oil was used as the odour source. Then they would sniff at the box only two or three times, before continuing on their way. The reason for these differing reactions has to do with the nature of these odour sources. The tiger quolls sniffed at the eucalyptus oil and the untreated platform as part of their normal investigative activities. Throughout each trial tiger quolls would spend most of the 25 minutes investigating and sniffing various surfaces within the cage, just as Green and Scarborough (1990) described. They attempted to extract the fish from the box, as the odour was indicative of the actual presence of the fish. But the urinary and faecal odours of bush rats and antechinus indicated an increased likelihood of encountering these prey species, rather than their actual presence.

Odour therefore probably plays a dual role for mammalian predators such as quolls which take an integrated approach to locating prey, being utilised in both the initial and final stages of predation. A number of studies have shown that the location of prey is significantly slower if any of the senses are unavailable (Wells 1978; Wells and Lehner

1978; Langley 1983; Langley 1985; Huang 1986; Langley 1988). But to thoroughly investigate every patch encountered to the fullest extent of their senses would be inefficient, as not every patch will contain food. Olfactory cues resulting from deliberate scent marks or residual integumentary odours are probably essential for scent hunting predators that employ a general “search strategy”; a basic set of rules of scanning and locomotion which results in the effective encountering of a specific distribution of prey (Smith 1974). Upon encountering olfactory cues indicative of the presence of prey species they adopt a more specific “search tactic”; a further adaptive change in scanning or locomotion increasing the likelihood of detecting the prey (Smith

1974). For example, Eurasian kestrels Falco tinnunculus are known to concentrate their Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 201

predatory efforts in an area upon detecting high levels of field vole urine, the distinctive ultraviolet spectral properties of which they are able to see (Viitala et al. 1995).

The responses to the urinary and faecal odours of some of their prey by tiger quolls in this study, and weasels in previous studies (Cushing 1984; Ylönen et al. 2003) indicate that these species may respond similarly to the scent marks of prey species. It is likely that while foraging these mammalian predators search for olfactory cues such as urine and faeces, as well as others such as hair, other scent marks, and even blood, to determine where to concentrate their search. If they detect odour trails resulting from deliberate scent marks or residual integumentary odours, they no doubt follow them to areas where a high concentration of these olfactory cues indicates an increased likelihood of encountering prey. Banks et al. (2000) predicted that animals with very low mobility were more vulnerable to predation, due to more time spent around a greater accumulation of such scent marks, and found this was indeed the case. Similarly the addition of extra urine and faeces to areas of field vole and sibling vole M. rossianmeridionalis habitat resulted in increased predation on these species by mustelids (Koivula and Korpimäki 2001).

It is possible that prey species alter their scent marking behaviour during periods of increased predation risk, in response to the utilisation of their scent marks as olfactory cues by predators. Roberts (2001) found that dominant male house mice Mus domesticus reduced their scent marking in response to ferret M. putorius furo urine.

Although scent marking has not been studied in bush rats and brown antechinus, swamp rats R. lutreolus may deposit over 150 urinary scent marks in an hour, which is comparable to the rates for house mice and bank voles (Mallick 1992). It would be Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 202

interesting to investigate whether these native species reduce their rate of scent marking in response to the odour of the tiger quoll.

There were several experimental issues which need to be considered when interpreting these results. Several quolls did not investigate both platforms, and had to be removed from the analyses, thereby affecting the power of the subsequent tests. However, where null hypotheses were accepted, there was little indication that additional replicates would have changed the outcome. In one case, the T-test conducted on the remaining five individuals still yielded a significant result. In all other cases the remaining results were equally distributed between the two platforms, and the fact that some of the quolls did not approach the platform at all is further evidence they either did not detect or were not attracted to the odour source.

Despite these considerations, the result from these experiments suggest that quolls do use odour in their predatory explorations, but not necessarily over as great a distance as has previously been assumed. I have argued that while predators may be able to detect odours over longer distances, this can only occur if the wind is strong enough and blowing in the right direction. Hence, for most predators the greatest application of their increased olfactory capabilities is likely to lie in their ability to differentiate and deduce information from the odour sources they encounter. This information then directs the way in which all the senses are utilised in an integrated and systematic search for prey. Greater recognition and incorporation of this concept into our consideration of predator/prey dynamics already has, and will continue to provide greater understanding about the way these species interact. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 203

8.6 Acknowledgements

I would like to thank the crew in the School of BEES workshop for constructing the experimental apparatus. Thanks also to Jonathan Russell for allowing me to borrow his

Handycam. And a massive thank you to Chad Staples from Featherdale Wildlife Park for organising access to the animals and the experimental cage, for help in transferring and handling the animals, and for his advice and support during the course of this experiment. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 204

8.7 References

Alexander, D. (1980). Diet of the Tiger Quoll (Dasyurus maculatus Kerr) in an area of New England National Park, Northeastern New South Wales. Honours thesis, University of New England, Armidale.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2000). Nonlinearity in the predation risk of prey mobility. Proceedings of the Royal Society of London: Series B, Biological Sciences. 267, 1621-1625.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2002). Mobility decisions and the predation risks of reintroduction. Biological Conservation 103, 133-138.

Bearder, S. K. (1977). Feeding habits of spotted hyaenas in a woodland habitat. East African Wildlife Journal 15, 263-280.

Belcher, C. A. (1995). Diet of the tiger quoll (Dasyurus maculatus) in East Gippsland, Victoria. Wildlife Research 22, 341-357.

Belcher, C. A. (1998). Susceptibility of the tiger quoll, Dasyurus maculatus, and the eastern quoll, Dasyurus viverrinus, to 1080-poisoned baits in control programmes for vertebrate pests in eastern Australia. Wildlife Research 25, 33- 40.

Burnett, S. E. (2000). Ecology and Conservation Status of the northern Spot-tailed Quoll, Dasyurus maculatus: with reference to the Future of Australia's Marsupial Carnivores. Ph.D. dissertation, James Cook University of Northern Queensland, Townsville: 292.

Cushing, B. S. (1984). A Selective Preference by Least Weasels for Oestrous Versus Dioestrous Urine of Prairie Deer Mice. Animal Behaviour 32, 1263-1265.

Edgar, R. and Belcher, C. A. (1998). Spotted Tailed quoll Dasyurus maculatus. In The Mammals of Australia. (edited by R. Strahan). New Holland Publishers, Sydney.

Endler, J. A. (1991). Interactions between predators and prey. In Behavioural Ecology: An Evolutionary Approach. (edited by J. R. Krebs and N. B. Davies) pp. 482. Blackwell Scientific Publications, Oxford.

Fleay, D. (1932). The rare dasyures (native cats). Victorian Naturalist 49, 63-69.

Glen, A. S. and Dickman, C. R. (2003). Effects of bait-station design on the uptake of baits by non-target animals during control programmes for foxes and wild dogs. Wildlife Research 30, 147-149.

Green, R. H. and Scarborough, T. J. (1990). The spotted-tailed quoll Dasyurus maculatus (Dasyuridae, Marsupialia) in Tasmania. The Tasmanian Naturalist 100, 1-15.

Holling, C. S. (1958). Sensory Stimuli involved in the location and selection of sawfly cocoons by small mammals. Canadian Journal of Zoology 36, 633-653. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 205

Huang, C. (1986). Detection of prey by captive Ningaui ridei and Sminthopsis griseoventer (Marsupialia: Dasyuridae). Australian Mammalogy 10, 23-26.

Johnson, R. P. (1975). Scent marking with urine in two races of the bank vole (Clethrionomys glareolus). Behaviour 55, 81-93.

Kats, L. B. and Dill, L. M. (1998). The scent of death: Chemosensory assessment of predation risk by prey animals. Ecoscience 5, 361-394.

Koivula, M. and Korpimäki, E. (2001). Do scent marks increase predation risk of microtine rodents? Oikos 95, 275-281.

Kortner, G., Gresser, S. and Harden, B. (2003). Does fox baiting threaten the spotted- tailed quoll, Dasyurus maculatus? Wildlife Research 30, 111-118.

Langley, W. M. (1983). Relative importance of the distance senses in grasshopper mouse predatory behaviour. Animal Behaviour 31, 199-205.

Langley, W. M. (1985). Relative importance of distance senses in hamster predatory behavior. Behavioural Processes 10, 229-239.

Langley, W. M. (1988). Spiny Mouse's (Acomys cahirinus) use of its distance senses in prey localization. Behavioural Processes 16, 67-73.

Lima, S. L. (2002). Putting predators back into behavioral predator-prey interactions. Trends in Ecology and Evolution 17, 70-75.

Lima, S. L. and Dill, L. M. (1990). Behavioral decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology 68, 619-640.

Macdonald, D. W. and Brown, R. E. (1985). Introduction: The pheromone concept in mammalian chemical communication. In Social odours in mammals. (edited by D. W. Macdonald and R. E. Brown). 1 pp. 1-36. Clarendon Press, Oxford.

Mallick, S. A. (1992). Urine-marking in Three Species of Rattus. Wildlife Research 19, 89-93.

Mech, L. D. (1970). The Wolf: the ecology and behavior of an endangered species. Natural History Press, New York.

Mills, M. G. L. (1978). Foraging Behaviour of the Brown Hyaena (Hyaena brunnea Thunberg, 1820) in the Southern Kalahari. Zeitshrift fuer Tierpsychologie 48, 113-141.

Minitab (2003). Minitab Release 14.1, Statistical Software. Pennsylvania, Minitab.

Murray, A. J. and Poore, R. N. (2004). Potential impacts of aerial baiting for wild dogs on a population of spotted-tailed quolls (Dasyurus maculatus). Wildlife Research 31, 639-644.

Österholm, H. (1964). The significance of distance receptors in the feeding behaviour of the fox (Vulpes vulpes L.). Acta Zoologica Fennica 106, 1-31. Chapter 8: Role of odour in the prey searching behaviour of tiger quolls 206

Raber, H. (1944). Versuche zur Ermittlung des Beuteschemas an einem Hausmarder (Martes foina) und einem Iltis (Putorius putorius). Revue Suisse De Zoologie 51, 293-332.

Roberts, S. C., Gosling, L. M., Thornton, E. A. and McClung, J. (2001). Scent-marking by male mice under the risk of predation. Behavioral Ecology 12, 698-705.

Rozenfeld, F. M., Le Boulange, E. and Rasmont, R. (1987). Urine marking by male bank voles (Clethrionomys glareolus Shreber, 1780: Microtidae, Rodentia) in relation to their social rank. Canadian Journal of Zoology 65, 2594-2601.

Smith, J. N. M. (1974). The food searching behaviour of two European Thrushes. II: The adaptiveness of the search patterns. Behaviour 49, 1-61.

Spencer, W. D. and Zielinski, W. J. (1983). Predatory Behavior of Pine Martens. Journal of Mammalogy 64, 715-717.

Viitala, J., Korpimäki, E., Palokangas, P. and Koivula, M. (1995). Attraction of kestrels to vole scent marks visible in ultraviolet light. Nature 373, 425-427.

Wells, M. C. (1978). Coyote senses in predation: environmental influences on their relative use. Behavioural Processes 3, 149-158.

Wells, M. C. and Lehner, P. N. (1978). The relative importance of the distance senses in coyote predatory behaviour. Animal Behaviour 26, 251-258.

Ylönen, H., Sundell, J., Tiilikainen, R., Eccard, J. A. and Horne, T. (2003). Weasels' (Mustela nivalis nivalis) preference for olfactory cues of the vole (Clethrionomys glareolus). Ecology 84, 1447-1452. Chapter 9:Conclusions and Implications 207

Chapter 9:

Conclusions and Implications

9.1 Summary of results

From the work reported in this thesis, it is clear that odour plays an important role in

Australian mammalian predator/prey interactions. In summary, native species responded to olfactory cues in the following ways (also see Table 9.1):

9.1.1 Native Rodents

Bush rats Rattus fuscipes, swamp rats R. lutreolus and eastern chestnut mice Pseudomys gracilicaudatus all avoided entering traps scented with the faeces of the native marsupial predator, the tiger quoll Dasyurus maculatus and the introduced placental predator, the red fox Vulpes vulpes (Chapter 4). A more detailed examination of the movement patterns of bush rats and swamp rats in a captive experiment showed that both species responded to both predator odours in the same manner, by decreasing their speed of movement in the enclosure where the odour was presented (Chapter 5).

9.1.2 Brown Antechinus

The marsupial brown antechinus Antechinus stuartii did not avoid entering traps scented with the faeces of tiger quolls and red foxes (Chapter 4). However, in captive experiments they increased their giving up density (GUD) in response to both fox and quoll odour and altered their movement patterns in response to the odour of tiger quoll faeces but not red fox faeces, by spending less time in the enclosure where the odour was presented (Chapter 5). Chapter 9:Conclusions and Implications 208

9.1.3 Bandicoots

Like the brown antechinus, long-nosed bandicoot Perameles nasuta and southern brown bandicoots Isoodon obesulus did not avoid entering fox-scented and quoll- scented traps, whereas northern brown bandicoots I. macrourus entered traps scented with tiger quoll faeces significantly more often than would happen by chance (Chapter

3). When their behaviour patterns were examined in a captive situation, long-nosed bandicoots and northern brown bandicoots increased their GUDs in response to both tiger quoll and red fox faecal odour. Both bandicoot species also decreased the amount of time they spent in the enclosures where quoll faeces were presented, although this response was stronger for the long-nosed bandicoot than the northern brown bandicoot.

However, northern brown bandicoots also decreased their average speed in response to the faecal odour of both predators (Chapter 6).

9.1.4 Common Brushtail Possums

As for northern brown bandicoots, common brushtail possums Trichosurus vulpecula entered traps scented with quoll faeces significantly more often than would happen by chance (Chapter 3) but their behaviour was not studied further here.

9.1.5 Tiger Quolls

Tiger quolls were attracted to the urine and faeces of bush rats and brown antechinus.

They followed urinary and faecal scent trails, and spent significantly more time in the area where the urine and faeces were presented, but appeared to be unable to detect these odour sources from distances exceeding 30 cm (Chapter 8). Chapter 9:Conclusions and Implications 209

Table 9.1: Summary of the responses of eight Australian prey species to the faecal odour of two predators, the native tiger quoll and the introduced red fox. Where - = no change, p = decrease, and n = increase. N/A indicates this response was not investigated for these species. Trappability Mobility Foraging Space Use Bush Rat Quoll pp-- Fox pp-- Swamp Rat Quoll pp-- Fox pp-- Eastern Chestnut Mouse Quoll p N/A N/A N/A Fox p N/A N/A N/A Brown Antechinus Quoll - - pp Fox - - p - Long-nosed Bandicoot Quoll - - pp Fox - - p - Northern Brown Bandicoot Quoll nppp Fox - ppn Southern Brown Bandicoot Quoll - N/A N/A N/A Fox - N/A N/A N/A Common Brushtail Possum Quoll n N/A N/A N/A Fox - N/A N/A N/A Chapter 9:Conclusions and Implications 210

9.2 Synthesis of prey responses to predator odours

These results represent the first demonstration that native Australian mammal species respond to the odour of a native Australian predator. Seven of the eight prey species tested, altered their behaviour in response to tiger quoll faecal odour in a manner likely to reduce their risk of predation. The eighth species, the southern brown bandicoot, was not trapped in high enough numbers to make any definitive conclusions. Six of these seven species also responded to the faecal odour of the introduced red fox in a manner likely to reduce their risk of predation. However, the marsupials did not appear to respond to fox odour as strongly as quoll odour, while the rodent responses to the two predator odours were very similar. The seventh species, the common brushtail possum, has previously been shown to respond to fox odour (Gresser 1996).

Common responses to native and introduced predators were not due to common components in the faecal odours of the two predator species. In Chapter 7, I found that the only common odour components in red fox and tiger quoll faeces are also common to the faecal odour of a number of non-predatory species. They are also unlikely to be responding to unfamiliar chemicals in red fox odour just because they are sulfur based, as prey species which have been shown to respond to sulfur-based compounds derived from predator urine and faeces do not respond to structurally similar sulfur-based compounds, even when derived from the same source odour (Sullivan and Crump 1984;

Vernet-Maury et al. 1984; Sullivan and Crump 1986).

An alternate hypothesis is that behavioural responses to foxes have evolved in some native wildlife after less than 150 years of contact. Predation pressure is a strong selective agent on phenotypic variants in a number of species (Hoekstra et al. 2004), Chapter 9:Conclusions and Implications 211

and for example, can lead to rapid changes in pelage colouration (Brown 1965). But if this is the case, why do rodents appear to respond more strongly to fox odour than marsupials, despite being in contact with them for the same period of time?

It is possible that initially rodents were under greater predation pressure from foxes than marsupial prey species. Marsupials would have been completely novel to foxes as potential prey, while murid rodents are a major component in the diet of British foxes

(Lever 1959; Richards 1977; Hewson 1983) and the morphological distinctions between foxes and other canids are thought to have evolved to make foxes especially efficient hunters of rodents and lagomorphs (Henry 1980; Henry 1996). A comparison of fox predation pressure on marsupials and rodents is difficult given the current scarcity of both groups, but Mitchell and Banks (2005) review of fox diets reported similar levels of predation on marsupial antechinus and bush rats (10.8% and 11.5% of fox scats respectively). The rapid decline of a suite of marsupial species, which is widely attributed to the arrival of the fox (Watts 1969; Christensen 1980; Southgate

1990; Short 1998; Richards and Short 2003), also indicates that foxes rapidly adapted to preying on these novel prey species and likely induced comparable selective pressure on marsupial and rodent prey.

Instead, different responses to foxes by marsupials and rodents may be due to their evolutionary histories. The ancestors of Australia’s native rodents evolved in southeast

Asia (Watts and Baverstock 1995), presumably alongside Asian canid predators

(Martin 1989; Wang et al. 2004), where avoidance of canid odours is likely to have evolved. The evolution of responses to predator odours has two components; the ability to recognise the odour, and the behavioural response once the odour is recognised. It Chapter 9:Conclusions and Implications 212

has recently been discovered that the recognition of odorants has a genetic basis (Buck and Axel 1991; Read 2004). The odorant receptor (OR) gene family is rapidly evolving

(Young and Trask 2002), and predation pressure is thought to be one of the main drivers of OR sequence diversification and gene duplication (Emes et al. 2004).

However, this process is occurring over a time scale of thousands of years (Young and

Trask 2002). Although OR genes are constantly being lost from the Genome, rodents appear to have relatively high rate of retention in comparison to humans (Young et al.

2002). If the OR genes for the recognition of common components of canid odours were retained within the genome of Australia’s native rodents, this would have allowed for a far more rapid evolution of responses to these odours than in species which had never encountered them before. This hypothesis would also explain why, over the same time period, house mice Mus domesticus have not evolved to respond to the odour of the western quoll D. geoffroii to the same degree as red fox and domestic cat Felis catus odour (Dickman 1992); prior to their arrival in Australia, house mice had no experience with any marsupial carnivores and their unique odour. It would also explain why native rodents appear to have evolved to respond to red fox odour faster than the marsupial species, although marsupials have been in contact with another canid, the dingo Canis lupus dingo, for 3500 to 4000 years (Corbett 1995).

While this may seem to suggest that after 4000 years of contact marsupials still haven’t evolved to respond to dingo odour to the same degree as tiger quoll odour, this is not necessarily the case. Over this time, marsupials may have evolved to respond to components in dingo odour which are not present in red fox odour. However, if rodents have retained a genetic recognition of components of canid odour, it is likely they are responding to a common chemical such as 3-methyl-3-butenyl methyl sulfide. A Chapter 9:Conclusions and Implications 213

number of different murid species have been shown to react adversely specifically to this volatile compounds including black rats Rattus rattus (Burwash et al. 1998), brown rats R. norvegicus (Bramley et al. 2000) and montane voles Microtus montanus and meadow voles M. pennsylvanicus (Sullivan et al. 1988). It is found in the urine of the wolf C. lupus (Raymer et al. 1984; Raymer et al. 1986), coyote C. latrans (Schultz et al. 1988), and domestic dog C. lupus familiaris (Schultz et al. 1985) and the urine and faeces of the red fox (Jorgenson et al. 1978; Wilson et al. 1978; Vernet-Maury et al.

1984), and is likely to also be present in the urine and faeces of the dingo.

Little is known of how native prey species respond to dingo odour. Blumstein et al.

(2002b) reported no change in the feeding behaviour of tammar wallabies Macropus tammar and red-necked pademelons Thylogale thetis in response to predator odours including dingo faeces. This is the only published study to specifically use dingo faeces, and they pooled their data such that the response to dingo faeces was not analysed separately to red fox and brown bear Ursus arctos faeces. However, there have been a couple of studies on the responses of native mammals to the odour of the closely related domestic dog. Banks et al. (2003) found that the trap success of bush rats was unaffected by dog faeces. In contrast, the application of dog urine reduced browsing on seedlings by swamp wallabies Wallabia bicolour (Montague et al. 1990), but this may have been due to reduced palatability rather than a response to the odour of the urine (Jones and Dayan 2000). More work must be conducted on the responses of

Australian native species to dingo odour, before it can be determined how this relates to their responses to red fox and tiger quoll odour. In particular it would be interesting to investigate the responses of Tasmanian prey species to canid odours, where dingoes have never been present. Chapter 9:Conclusions and Implications 214

9.3 Scalar effects on odour signal exploitation

This thesis also showed for the first time that tiger quolls respond to olfactory cues from prey species. Tiger quolls readily followed scent trails once detected, and followed them back to their source odours, spending more time in these areas than identical unscented areas. However, the distance over which they could detect these odours was less than 65 cm. Anecdotal reports estimate brown hyenas Hyaena brunnea can detect carrion by its odour from 1.9 km away (Mills 1978), and wolves can detect moose from 2.4 km away (Mech 1970). But these accounts were exceptions and due to strong winds. In such conditions the odour source was only detected if these predators were downwind, and was not detected upwind even at very small distances (Mech

1970; Mills 1978). Indeed, in wind speeds of less than 10 km/h the maximum range over which mammalian predators are able to detect an odour source appears to be less than 1.5 m (Table 9.2).

In still air conditions, odours diffuse at a slow but equal rate in all directions away from the source, and the range at which the concentration remains above that necessary to be detected by vertebrate olfaction is limited to this distance of approximately 1.5 m

(Bossert and Wilson 1963). A strong wind may increase the range at which the odour is above this concentration by several hundred metres or more (Bossert and Wilson 1963), however this also increases the large scale (David et al. 1982; David et al. 1983;

Elkinton et al. 1987) and small scale (Murlis and Jones 1981; Murlis et al. 2000) fluctuations within the odour plume created. Due to this increased variability, although animals may become aware of an odour that has an origin tens or hundreds of metres upwind of their current position, more often than not, they are unable to successfully Chapter 9:Conclusions and Implications 215

track these odour plumes back to their source (Elkinton et al. 1987) (see Appendix 1 for more detail).

Predators therefore probably use olfaction most effectively to detect trace odours of prey species at close range. Detection of such trace odours, be they derived from skin, hair, blood, urine, faeces or other scent markings during the course of a predator’s normal investigative “search strategy” may lead to the adoption of a “search tactic”

(Smith 1974) such as “area-restricted searching” (Tinbergen et al. 1967; Macdonald

1980). If this is the case, it has important implications in terms of why accumulations of scent marks are dangerous to prey species. For example, Koivula and Korpimäki

(2001), found higher mortality of voles from scent hunting weasels in areas treated with additional vole urine, even though there was no increase in weasel numbers in these treated patches. It is likely that mustelids, which would have passed through the area anyway may have stopped and spent more time searching these areas for voles, resulting in the higher mortality. Similarly voles with low mobility may be more vulnerable to predation by mammalian scent-hunting predators (Banks et al. 2000;

Banks et al. 2002), not because these predators are attracted over long distances to accumulations of scent marks and waste products, but rather because they spend more time searching in an area when they encounter these odour sources.

Prey species may similarly be unable to detect predator odours over long distances in the absence of wind. Laboratory brown rats have to come within 30 cm of cat fur before they will respond to its odour (Blanchard et al. 2003), and for a previously worn cat collar they don’t respond until they are within 5 cm (Dielenberg and McGregor 2001).

Thus, individuals may also have to come into close proximity with a predator’s scent Chapter 9:Conclusions and Implications 216

mark before they respond to it, and may further investigate the area before abandoning it. This may explain seemingly dangerous behaviours such as deer species investigating the scent marks of tigers Panthera tigris at close range (Muller-Schwarze 1972). It may also explain why meadow voles spent more time sniffing weasel urine than the alternate guinea pig urine or the odourless control, but spent less time in total in the chamber where weasel odour was presented, yet didn’t appear to avoid weasel-scented traps in the field (Parsons and Bondrup-Nielsen 1996). In turn, such initial scent inspection may also be why brown antechinus and long-nosed bandicoots were trapped in equal numbers in quoll-scented traps and unscented traps in the field, but spent less time in the area where quoll scats were presented in a captive situation.

Tiger quolls do respond to olfactory cues from potential prey species, but in the absence of wind the range of detection of these odours is limited. However, this likely applies to many mammalian species; both predators and prey. Olfactory cues are still likely to play an important role in their detection of prey, and future investigations into how exactly they respond to these cues in the wild may lead to a much greater understanding of how odour is utilised in predator prey interactions. Chapter 9:Conclusions and Implications 217

Table 9.2: Maximum distance over which odour sources could be detected by mammalian predators in conditions with a wind speed of less than 10 km/h. Predator Odour Source Distance Study Red fox meat piece <1.5 m (Österholm 1964) Coyote rabbit Oryctolagus <1.5 m (Wells 1978) cuniculus Domestic Nesting red grouse Lagopus <0.5 m (Hudson 1992) dog lagopus Domestic Nesting black grouse T. <1.1 m (Storaas et al. dog tetrix 1999) Domestic Nesting capercaillie Tetrao <1.5 m (Storaas et al. dog urogallus 1999) Tiger quoll whiting <0.65 m Chapter 8 bush rat and antechinus urine and faeces

9.4 Olfaction vs other senses used in predator avoidance

Some authors have suggested that olfactory cues of predation risk may be generally less effective than visual and aural cues (Blumstein et al. 2002b; Jones et al. 2004). They argue that because odours persist long after the donor has departed, and that all carnivore scents could trigger anti predator responses rather than only those of potential predators, potential prey may make costly errors in estimating their predation risk

(Blumstein et al. 2002b; Jones et al. 2004). However, olfactory cues represent the recent presence of the predator, and indicate that there is likely to be a high level of predator activity in the area. This is not just a sign to prey individuals to be more vigilant, but also allows prey species to change their behaviour before coming into direct contact with the predator, in such a manner as leaving the area, using cover, Chapter 9:Conclusions and Implications 218

moving more slowly, etc. so that they reduce their probability of detection by the predator.

A number of papers have investigated the responses of native Australian prey species to the visual cues of the introduced placental predators the red fox and the feral cat and suggested that these animals can be trained to respond to such visual cues (McLean et al. 1996; Blumstein et al. 2000; Griffin et al. 2000; McLean et al. 2000; Blumstein et al. 2002a; Jose et al. 2002; Griffin and Evans 2003a; Griffin and Evans 2003b).

However, such training regimes have failed to increase the subsequent survival of these animals upon release into the wild (McLean et al. 1996). This failure may have been due to the unrealistic nature of the puppets and models or stuffed animals on wheeled cart used to simulate the predators, as real live predators are unlikely to move in a similar fashion, or it may have been because of the nature of the cue itself.

Endler (1991) describes six stages of predation; Encounter, Detection, Identification,

Approach, Subjugation, and Consumption. Prey should deploy anti-predator strategies as early as possible within this sequence for three inter-related reasons; more opportunities to deploy alternate strategies if the initial strategies fail, cost of anti- predator strategies increase at each stage, and actual risk of predation increases at each stage (Endler 1991). Olfactory cues allow prey species to respond to predation risk before actually encountering the predator, and so are likely to be more effective and less costly than responses to visual cues, which may not come into effect until the predator is already in the approach phase. While both cues are no doubt important, responses solely to visual cues may not be enough to significantly decrease predation, especially considering the predatory strategies of the red fox. Unlike other canids, Chapter 9:Conclusions and Implications 219

which tend to approach prey openly and run down weaker or more vulnerable animals, fox predation is largely based on stealth, stalking its prey to avoid detection, before pouncing on or rushing the prey once close enough (Henry 1980; Henry 1996). So by the time that native prey species respond to visual cues it may be too late to avoid predation.

9.5 Implications of olfactory signal exploitation to conservation

Red foxes have had a disastrous impact on Australian wildlife and they remain amongst the worst of the threatening processes which continue to put remnant fauna at risk of extinction. Short (2002) suggested that foxes had such a major impact on marsupials because they were more efficient and successful than native predators, and that these native species lacked effective anti-predator strategies against a novel predator of the size and hunting style of the fox. However, tiger quolls also stalk their prey before pouncing (Troughton 1943; Jones et al. 2001), may reach up to 7kg (Settle 1978), and in areas where they co-exist their diet is extremely similar to that of the red fox (Glen

2005). While there are clearly some differences between the predatory behaviours of quolls and foxes, to suggest that native species have never dealt with a predator of the same size or hunting style is unrealistic. And predation pressure by tiger quolls was great enough for anti-predator strategies to evolve. The results presented here suggest that marsupials may have been vulnerable to introduced predators such as the red fox, not because they didn’t have effective anti-predator strategies, but rather because they were naïve to the risk implied by olfactory cues from these predators, and failed to initiate their anti-predator strategies at the earlier stages of the predator/prey interaction. Chapter 9:Conclusions and Implications 220

Introduced predators have had a greater impact on native species in Australasia than in other parts of the world (Salo et al. In Review). This may be due to relative levels of naiveté of native prey species to introduced predators. Throughout most of the rest of the world, introduced carnivores have tended to be closely related to indigenous predators. For example, the American mink M. vison, which has become established in many part of Europe (Kauhala 1996), although larger in size, is otherwise very similar to the European mink M. lutreola, and is the same size as the also closely related polecat M. putorius (Macdonald and Barrett 1993). Therefore, alien predators tend not to be radically different in terms of morphology and behaviour from their native counterparts. Closely related species are also more likely to share common components in their odours. For example, there are a number of common odour components in the anal sac secretions of American mink and European mustelids (Brinck et al. 1983), which European prey species have been shown to respond to (Robinson 1990). In other cases, introduced carnivores may not be as closely related to their local counterparts, such as the introduction of the common raccoon Procyon lotor to Japan (Makoto et al.

2003), where no Procyonids existed previously (Macdonald 2001). However, raccoons may have some odour components in common with native Japanese mustelids, as the

Procyonidae are not that far removed from the Mustelidae evolutionarily speaking

(Macdonald 2001), and mustelids share some odour components with the more distantly related Canidae, which prey species are known to respond to (Sokolov et al.

1980; Brinck et al. 1983; Vernet-Maury et al. 1984; Sullivan and Crump 1986).

In contrast, marsupial predators are from an entirely different Infraclass to the placental carnivores (Macdonald 2001). Although superficially similar in terms of morphology and behaviour due to convergent evolution (Augee and Fox 2000), the differences Chapter 9:Conclusions and Implications 221

between marsupial and placental predators are greater than for similar species within the Carnivora. But beyond this, my results indicate that the chemical components of marsupial predator odours which prey species respond to are not present in placental carnivore odours. So, not only were Australian native species naïve to differences in predatory behaviour between native marsupial and introduced placental predators, they were also naïve to the increase in risk necessary to initiate their anti-predator strategies.

In New Zealand, where the only terrestrial native mammals prior to the arrival of human beings 1000 years ago (Flannery 1997) were bats (King 1998), the native fauna were completely naïve to any mammalian predator (Maloney and McLean 1995;

McLean et al. 1999), and the impact of introduced carnivores, such as the weasel, stoat

M. erminea, ferret M. putorius furo, and cat (King 1998), has arguably been even greater than in Australia (King 1984).

But the question needs to be asked, why were foxes able to adapt to hunting Australian prey faster than these native species were able to adapt to avoid predation? In terms of the predator/prey arms race (Dawkins and Krebs 1979), foxes would have been just as naïve to these prey species as they were to the fox. Possibly this is why several releases of foxes prior to the 1870s were unsuccessful (Rolls 1969). The main reason why foxes were able to spread after this time is likely due to the concurrent spread of the rabbit, the fox’s traditional prey from Europe (Jarman 1986; Saunders et al. 1995). After extreme persecution of quolls and other native predators, which were blamed for frustrating earlier efforts, rabbits became established in Victoria in the early 1860s and rapidly spread across the country (Rolls 1969; Caughley 1980). The spread of foxes after their initial establishment in Victoria in the early 1870s then closely followed that of the rabbit, sometimes with a lag of only a few years (Jarman 1986; Saunders et al. Chapter 9:Conclusions and Implications 222

1995). This tipped the balance of this inchoate system in favour of the fox. Failure to capture native prey was offset by an abundant supply of a species to which the fox had coevolved to hunt (Henry 1980; Henry 1996); failure to avoid fox predation by native species was lethal.

Without the rabbit, foxes would have been forced to rely on native prey species for food. Failure to successfully prey upon species which they had not evolved to hunt would have been far more detrimental, drastically reducing their rate of range expansion, and allowing native prey species time to adapt their anti-predator strategies to this new predator. This hypothesis may also explain why the dingo doesn’t appear to have had as drastic an impact on native species when it was introduced to Australia

4000 years ago (Morton 1990; Smith and Quin 1996; Short et al. 2002). Putting this hypothesis in the context of predator/prey theory; prior to the arrival of introduced predators, native prey species are more likely to have been experiencing a type III functional response from native predators, as anti-predator strategies would have significantly decreased the likelihood of predation at low population densities (Sinclair et al. 1998). However, prey which are completely naïve to introduced predators are more likely to display a type II response, and would be more likely to be driven to extinction (Sinclair et al. 1998) if they were secondary prey species and the predators were maintained at a high density due to a large population of the primary prey species

(Pech et al. 1995). This “hyperpredation” scenario, with foxes and rabbits, has been proposed as one of the primary reasons for the decline and extinction of so many native

Australian mammal species since European settlement (Smith and Quin 1996). The results of this thesis show that native species respond to olfactory cues of the tiger quoll and apply anti-predator strategies, and that they may be adapting these responses to the Chapter 9:Conclusions and Implications 223

olfactory cues of red foxes. This could help to shift the functional response from type II to type III, reducing the likelihood of further extinctions due to the red fox, although this is likely to also be dependent on other variables such as habitat (Sinclair et al.

1998).

This hypothesis should not be interpreted as saying that Australian wildlife are no longer at risk from fox predation; the adaptation to predation risk from foxes is likely to be occurring at different rates for different species, and some species may remain more vulnerable than others. For example, broad-toothed rats Mastocomys fuscus are more vulnerable to predation by foxes than bush rats in the Snowy Mountains of Australia

(Green 2002). It is also possible that some anti-predator behaviours involving olfaction, which have developed to reduce predation by quolls, are ineffective against foxes. In subalpine areas predation by foxes on broad-toothed rats is particularly high during the winter months (Green 2002), when broad-toothed rats nest communally in subnivean nests (Bubela and Happold 1993). These nests may be detectable by foxes, which can detect meat pieces 10 cm below the ground (Österholm 1964), but not by tiger quolls, which could not detect meat bait buried 5 cm below the surface of the ground, even when placed in obviously disturbed bait stations (Belcher 1998). Similarly, the nests built by the northern brown bandicoots and long-nosed bandicoots used in Chapter 6 were all approximately 8 cm below the surface of the ground, beyond the range of detection of tiger quolls, but within the range of foxes. Further work on the exploitation of odours by these two predators is clearly needed to understand this other role of odour in predator prey interactions. The strategy of northern brown bandicoots to stay still if their nests are disturbed may be ineffective if foxes can detect them at this depth, and they may be more vulnerable to foxes than long-nosed bandicoots, which bolt from the Chapter 9:Conclusions and Implications 224

nest at the slightest sign of disturbance. It would be interesting to determine which of these strategies other bandicoot species have developed, and whether this affects their vulnerability to fox predation.

In light of the importance of odour in mammalian predator/prey interactions revealed in this thesis, several lines of research are needed to understand some of the more complex aspects of these interactions. In addition to work suggested at the end of sections 9.2 and 9.3, future studies should also investigate the range of responses prey species have to both predators. As anti-predator strategies may result in costs such as reductions in opportunities to feed, breed and socialise (Lima 1998), the sub-lethal impacts of anti- predator behaviours shown by native species may also have considerable conservation implications. A more thorough investigation of the hunting habits of the two predators will also allow prey responses to be put into a more realistic context (Lima 2002). By better understanding and comparing how predators and prey are responding to each other, we may be able to determine which species are particularly vulnerable to predation and implement management strategies to lessen their vulnerability. Thus, a better understanding of these predator/prey interactions will lead to a greater ability to successfully protect Australia’s unique mammalian fauna. Chapter 9:Conclusions and Implications 225

9.6 References

Augee, M. and Fox, M. (2000). Biology of Australia and New Zealand. Pearson Education, Frenchs Forest.

Banks, P. B., Hughes, N. K. and Rose, T. A. (2003). Do native Australian small mammals avoid faeces of domestic dogs? Responses of Rattus fuscipes and Antechinus stuartii. Australian Zoologist 32, 406-409.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2000). Nonlinearity in the predation risk of prey mobility. Proceedings of the Royal Society of London: Series B, Biological Sciences. 267, 1621-1625.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2002). Mobility decisions and the predation risks of reintroduction. Biological Conservation 103, 133-138.

Belcher, C. A. (1998). Susceptibility of the tiger quoll, Dasyurus maculatus, and the eastern quoll, Dasyurus viverrinus, to 1080-poisoned baits in control programmes for vertebrate pests in eastern Australia. Wildlife Research 25, 33- 40.

Blanchard, D. C., Markham, C., Yang, M., Hubbard, D., Madarang, E. and Blanchard, R. J. (2003). Failure to Produce Conditioning With Low-Dose Trimethylthiazoline or Cat Feces as Unconditioned Stimuli. Behavioral Neuroscience 117, 360-368.

Blumstein, D. T., Daniel, J. C., Griffin, A. S. and Evans, C. S. (2000). Insular tammar wallabies (Macropus eugenii) respond to visual but not acoustic cues from predators. Behavioral Ecology 11, 528-535.

Blumstein, D. T., Daniel, J. C., Schnell, M. R., Ardon, J. G. and Evans, C. S. (2002a). Antipredator behaviour of red-necked pademelons: a factor contributing to species survival? Animal Conservation 5, 325-331.

Blumstein, D. T., Mari, M., Daniel, J. C., Ardon, J. G., Griffith, A. S. and Evans, C. S. (2002b). Olfactory predator recognition: wallabies may have to learn to be wary. Animal Conservation 50, 143-198.

Bossert, W. H. and Wilson, E. O. (1963). The Analysis of Olfactory Communication Among Animals. Journal of Theoretical Biology 5, 443-469.

Bramley, G. N., Waas, J. R. and Henderson, H. V. (2000). Responses of wild norway rats (Rattus norvegicus) to predator odors. Journal of Chemical Ecology 26, 705-719.

Brinck, C., Erlinge, S. and Sandell, M. (1983). Anal sac secretion in Mustelids: A comparison. Journal of Chemical Ecology 9, 727-745.

Brown, L. N. (1965). Selection in populations of house mice containing mutant individuals. Journal of Mammalogy 46, 461-465. Chapter 9:Conclusions and Implications 226

Bubela, T. M. and Happold, D. C. D. (1993). The Social Organisation and Mating System of an Australian Subalpine Rodent, the Broad-toothed Rat, Mastacomys fuscus Thomas. Wildlife Research 20, 405-417.

Buck, L. and Axel, R. (1991). A Novel Mutligene Family May Encode Odorant Receptors: A Molecular Basis for Odor Recognition. Cell 65, 175-187.

Burwash, M. D., Tobin, M. E., Woolhouse, A. D. and Sullivan, T. P. (1998). Laboratory evaluation of predator odors for eliciting an avoidance response in roof rats (Rattus rattus). Journal of Chemical Ecology 24, 49-66.

Caughley, J. (1980). Native Quolls and Tiger Quolls. In Endangered Animals of NSW. (edited by C. Haigh) pp. 45-48. NSW National Parks and Wildlife Service, Sydney.

Christensen, P. E. S. (1980). A sad day for native fauna. Forest Focus 23, 3-12.

Corbett, L. (1995). The Dingo in Australia and Asia. UNSW Press, Sydney.

David, C. T., Kennedy, J. S. and Ludlow, A. R. (1983). Finding of a sex pheromone source by gypsy moths released in the field. Nature 303, 804-806.

David, C. T., Kennedy, J. S., Ludlow, A. R., Perry, J. N. and Wall, C. (1982). A reappraisal of insect flight towards a distant point source of wind-borne odor. Journal of Chemical Ecology 8, 1207-1215.

Dawkins, R. and Krebs, J. R. (1979). Arms races between and within species. Proceedings of the Royal Society of London B 205, 489-511.

Dickman, C. R. (1992). Predation and Habitat Shift in the House Mouse, Mus domesticus. Ecology 73, 313-322.

Dielenberg, R. A. and McGregor, I. S. (2001). Defensive behavior in rats towards predatory odors: a review. Neuroscience and Biobehavioral Reviews 25, 597- 609.

Elkinton, J. S., Schal, C., Ono, T. and Carde, R. T. (1987). Pheromone puff trajectory and upwind flight of male gypsy moths in a forest. Physiological Entomology 12, 399-406.

Emes, R. D., Beatson, S. A., Ponting, C. P. and Goodstadt, L. (2004). Evolution and Comparative Genomics of Odorant- and Pheromone-Associated Genes in Rodents. Genome Research 14, 591-602.

Endler, J. A. (1991). Interactions between predators and prey. In Behavioural Ecology: An Evolutionary Approach. (edited by J. R. Krebs and N. B. Davies) pp. 482. Blackwell Scientific Publications, Oxford.

Flannery, T. F. (1997). The Future Eaters. New Holland Publishers, Sydney. Chapter 9:Conclusions and Implications 227

Glen, A. S. (2005). Do introduced predators compete with native carnivores? Proceedings of the 13th Australasian Vertebrate Pest Conference, Wellington, VPC.

Green, K. (2002). Selective predation on the broad-toothed rat, Mastocomys fuscus (Rodentia: Muridae), by the introduced red fox, Vulpes vulpes (Carnivora: Canidae), in the Snowy Mountains, Australia. Austral Ecology 27, 353-359.

Gresser, S. (1996). Anti-predator Behaviour of the Common Brushtail Possum (Trichosurus vulpecula) at Burrendong Dam, N.S.W. Honours Thesis, University of Sydney, Sydney: 106.

Griffin, A. S., Blumstein, D. T. and Evans, C. S. (2000). Training Captive-Bred or Translocated Animals to Avoid Predators. Conservation Biology 14, 1317-1326.

Griffin, A. S. and Evans, C. S. (2003a). The role of differential reinforcement in predator avoidance learning. Behavioural Processes 61, 87-94.

Griffin, A. S. and Evans, C. S. (2003b). Social learning of antipredator behaviour in a marsupial. Animal Behaviour 66, 485-492.

Henry, J. D. (1980). Fox Hunting. Natural History 89, 60-69.

Henry, J. D. (1996). Red Fox: The Catlike Canine. Smithsonian Institute Press, Washington D.C.

Hewson, R. (1983). The food of wild cats (Felis silvestris) and red foxes (Vulpes vulpes) in west and north-east Scotland. Journal of Zoology (London) 200, 283- 289.

Hoekstra, H. E., Drumm, K. E. and Nachman, M. W. (2004). Ecological genetics of adaptive colour polymorphism in pocket mice: geographic variation in selected and neutral genes. Evolution 58, 1329-1341.

Hudson, P. J. (1992). Grouse in Space and Time: The Population Biology of a Managed Gamebird. Game Conservancy Limited, Fordingbridge.

Jarman, P. (1986). The red fox - an exotic, large predator. In The Ecology of Exotic Animals and Plants: some Australian Case Histories. (edited by R. L. Kitching) pp. 43-61. John Wiley and Sons, Brisbane.

Jones, M. and Dayan, T. (2000). Foraging behaviour and microhabitat use by spiny mice Acomys cahirinus and A. russatus, in the presence of Blandford's fox (Vulpes vulpes) odor. Journal of Chemical Ecology 26, 455-469.

Jones, M. E., Rose, R. K. and Burnett, S. (2001). Dasyurus maculatus. Mammalian Species 676, 1-9.

Jones, M. E., Smith, G. C. and Jones, S. M. (2004). Is anti-predator behaviour in Tasmanian eastern quolls (Dasyurus viverrinus) effective against introduced predators? Animal Conservation 7, 155-160. Chapter 9:Conclusions and Implications 228

Jorgenson, J. W., Novotny, M., Carmack, M., Copeland, G. B., Wilson, S. R., Katona, S. and Whitten, W. K. (1978). Chemical scent constituents in the urine of the red fox (Vulpes vulpes) during the winter season. Science 199, 796-798.

Jose, D. P., McLean, I. G. and Calver, M. C. (2002). Recognition by two species of captive-reared peramelids. Australian Mammalogy 24, 213-217.

Kauhala, K. (1996). Distributional history of the American mink (Mustela vison) in Finland with special reference to the trends in the otter (Lutra lutra) populations. Annales Zooloici Fennici 33, 283-291.

King, C. (1984). Immigrant Killers: Introduced predators and the conservation of birds in New Zealand. Oxford University Press, Auckland.

King, C. M. (1998). The Handbook of New Zealand Mammals. Oxford University Press, Auckland.

Koivula, M. and Korpimäki, E. (2001). Do scent marks increase predation risk of microtine rodents? Oikos 95, 275-281.

Lever, R. J. A. (1959). The diet of the fox since myxomatosis. Journal of Animal Ecology 28, 359-375.

Lima, S. L. (1998). Nonlethal Effects in the Ecology of Predator-Prey Interactions: What are the ecological effects of anti-predator decision-making? Bioscience 48, 25-34.

Lima, S. L. (2002). Putting predators back into behavioral predator-prey interactions. Trends in Ecology and Evolution 17, 70-75.

Macdonald, D. W. (1980). The Red Fox, Vulpes vulpes, as a predator upon Earthworms, Lumbricus terrestris. Zeitshrift fuer Tierpsychologie 52, 171-200.

Macdonald, D. W. (2001). The New Encyclopedia of Mammals. Oxford University Press, Oxford.

Macdonald, D. W. and Barrett, P. (1993). Collins Field Guide: Mammals of Britain and Europe. Harper Collins, London.

Makoto, A., Matoba, Y., Ikeda, T., Suzuki, M., Asakawa, M. and Ohtaishi, N. (2003). Reproductive Characterisitics of the Feral Raccoon (Procyon lotor) in Hokkaido, Japan. Journal of Veterinary Medical Science 65, 369-373.

Maloney, R. F. and McLean, I. G. (1995). Historical and experimental learned predator recognition in free-living New Zealand robins. Animal Behaviour 50, 1193- 1201.

Martin, L. D. (1989). Fossil History of the Terrestrial Carnivora. In Carnivore Behavior, Ecology, and Evolution. (edited by J. L. Gittleman). Cornell University Press, New York. Chapter 9:Conclusions and Implications 229

McLean, I. G., Holzer, C. and Studholme, B. J. S. (1999). Teaching predator- recognition to a naive bird: implications for management. Biological Conservation 87, 123-130.

McLean, I. G., Lundie-Jenkins, G. and Jarman, P. J. (1996). Teaching an endangered mammal to recognise predators. Biological Conservation 75, 51-62.

McLean, I. G., Schmitt, N. T., Jarman, P. J., Duncan, C. and Wynne, C. D. L. (2000). Learning for life: Training marsupials to recognise introduced predators. Behaviour 137, 1361-1376.

Mech, L. D. (1970). The Wolf: the ecology and behavior of an endangered species. Natural History Press, New York.

Mills, M. G. L. (1978). Foraging Behaviour of the Brown Hyaena (Hyaena brunnea Thunberg, 1820) in the Southern Kalahari. Zeitshrift fuer Tierpsychologie 48, 113-141.

Mitchell, B. D. and Banks, P. B. (2005). Do wild dogs exclude foxes? Evidence for competition from dietary and spatial overlaps. Austral Ecology 30, 581-591.

Montague, T. L., Pollock, D. C. and Wright, W. (1990). An examination of the Browsing animal problem in Australian eucalypt and pine plantations. Proceedings of Vertebrate Pest Conferences 14, 203-208.

Morton, S. R. (1990). The impact of European Settlement on the vertebrate animals of arid Australia: a conceptual model. Proceedings of the Ecological Society of Australia 16, 210-213.

Muller-Schwarze, D. (1972). Responses of young black-tailed deer to predator odors. Journal of Mammalogy 53, 393-394.

Murlis, J. and Jones, C. D. (1981). Fine-scale structure of odour plumes in relation to insect orientation to distant pheromone and other attractant sources. Physiological Entomology 6, 71-86.

Murlis, J., Willis, M. A. and Carde, R. T. (2000). Spatial and temporal structures of pheromone plumes in fields and forests. Physiological Entomology 25, 211-222.

Österholm, H. (1964). The significance of distance receptors in the feeding behaviour of the fox (Vulpes vulpes L.). Acta Zoologica Fennica 106, 1-31.

Parsons, G. J. and Bondrup-Nielsen, S. (1996). Experimental analysis of behaviour of meadow voles (Microtus pennsylvanicus) to odours of the short-tailed weasel (Mustela erminea). Ecoscience 3, 63-69.

Pech, R. P., Sinclair, A. R. E. and Newsome, A. E. (1995). Predation Models for Primary and Secondary Prey Species. Wildlife Research 22, 55-64.

Raymer, J., Wiesler, D., Novotny, M., Asa, C., Seal, U. S. and Mech, L. D. (1984). Volatile constituents of wolf (Canis lupus) urine as related to gender and season. Experientia 40, 707-709. Chapter 9:Conclusions and Implications 230

Raymer, J., Wiesler, D., Novotny, M., Asa, C., Seal, U. S. and Mech, L. D. (1986). Chemical scent constituents in urine of wolf (Canis lupus) and their dependence on reproductive hormones. Journal of Chemical Ecology 12, 297-314.

Read, R. R. (2004). After the Holy Grail: Establishing a Molecular Basis for Mammalian Olfaction. Cell 116, 329-336.

Richards, D. F. (1977). Observations on the diet of the red fox (Vulpes vulpes) in South Devon. Journal of Zoology (London) 183, 495-504.

Richards, J. D. and Short, J. (2003). Reintroduction and establishment of the western barred bandicoot Perameles bougainville (Marsupialia: Peramelidae) at Shark Bay, Western Australia. Biological Conservation 109, 181-195.

Robinson, I. (1990). The effect of mink odour on rabbits and small mammals. In Chemical Signals in Vertebrates V. (edited by D. W. Macdonald). Oxford University Press, New York.

Rolls, E. C. (1969). They All Ran Wild, the Animals and Plants that Plague Australia. Angus and Robertson, Sydney.

Salo, P., Korpimäki, E., Nordstrom, M., Banks, P. B. and Dickman, C. R. (In Review). Are alien predators more dangerous than native predators to ecosystem health?

Saunders, G., Coman, B., Kinnear, J. and Braysher, M. (1995). Managing Vertebrate Pests: Foxes. Australian Government Publishing Service, Canberra.

Schultz, T. H., Flath, R. A., Stern, D. J., Mon, T. R., Teranishi, R., Kruse, S. M., Butler, B. and Howard, W. E. (1988). Coyote Estrous Urine Volatiles. Journal of Chemical Ecology 14, 701-712.

Schultz, T. H., Kruse, S. M. and Flath, R. A. (1985). Some volatile components of female dog urine. Journal of Chemical Ecology 11, 169-175.

Settle, G. A. (1978). The quiddity of tiger quolls. Australian Natural History 19, 164- 169.

Short, J. (1998). The extinction of rat-kangaroos (Marsupialia: Potoroidae) in New South Wales, Australia. Biological Conservation 86, 365-377.

Short, J., Kinnear, J. E. and Robley, A. (2002). Surplus killing by introduced predators in Australia-evidence for ineffective anti-predator adaptations in native prey species? Biological Conservation 103, 283-301.

Sinclair, A. R. E., Pech, R. P., Dickman, C. R., Hik, D., Mahon, P. and Newsome, A. E. (1998). Predicting Effects of Predation on Conservation of Endangered Prey. Conservation Biology 12, 564-575.

Smith, A. P. and Quin, D. G. (1996). Patterns and causes of extinction and decline in Australian conilurine rodents. Biological Conservation 77, 243-267. Chapter 9:Conclusions and Implications 231

Smith, J. N. M. (1974). The food searching behaviour of two European Thrushes. II: The adaptiveness of the search patterns. Behaviour 49, 1-61.

Sokolov, V. E., Albone, E. S., Flood, P. F., Heap, P. F., Kagan, M. Z., Vasilieva, V. S., Roznov, V. V. and Zinkevich, E. P. (1980). Secretion and secretory tissues of the anal sac of the mink, Mustela vison. Journal of Chemical Ecology 6, 805- 825.

Southgate, R. I. (1990). Distribution and abundance of the Greater Bilby Macrotis lagotis Reid (Marsupialia: Peramelidae). In Bandicoots and Bilbies. (edited by J. H. Seebeck, P. R. Brown, R. L. Wallis and C. M. Kemper) pp. 293-302. Surrey Beatty and Sons, Sydney.

Storaas, T., Kastdalen, L. and Wegge, P. (1999). Detection of forest grouse by mammalian predators: A possible explanation for high brood losses in fragmented landscapes. Wildlife Biology 5, 187-192.

Sullivan, T. P. and Crump, D. R. (1984). Influence of mustelid scent-gland compounds on suppression of feeding by snowshoe hares (Lepus americanus). Journal of Chemical Ecology 10, 1809-1821.

Sullivan, T. P. and Crump, D. R. (1986). Feeding responses of snowshoe hares (Lepus americanus) to volatile constituents of red fox (Vulpes vulpes) urine. Journal of Chemical Ecology 12, 729-739.

Sullivan, T. P., Crump, D. R. and Sullivan, D. S. (1988). Use of predator odors as repellents to reduce feeding damage by herbivores III. Montane and Meadow Voles (Microtus montanus and Microtus pennsylvanicus). Journal of Chemical Ecology 14, 363-377.

Tinbergen, N., Impekoven, M. and Franck, D. (1967). An experiment on spacing-out as a defence against predation. Behaviour 28, 307-321.

Troughton, E. (1943). Furred animals of Australia. Angus and Robertson, Sydney.

Vernet-Maury, E., Polak, E. H. and Demael, A. (1984). Structure-activity relationship of stress-inducing odorants in the rat. Journal of Chemical Ecology 10, 1007- 1018.

Wang, X., Tedford, R. H., Van Valenburgh, B. and Wayne, R. K. (2004). Ancestry: Evolutionary history, molecular systematics, and evolutionary ecology of the Canidae. In Biology and Conservation of Wild Canids. (edited by D. W. Macdonald and C. Siller-Zubiri) pp. 39-54. Oxford University Press, New York.

Watts, C. H. S. (1969). Distribution and Habits of the Rabbit Bandicoot. Transactions of the Royal Society of South Australia 93, 135-141.

Watts, C. H. S. and Baverstock, P. R. (1995). Evolution in the Murinae (Rodentia) assessed by Microcomplement Fixation of Albumin. Australian Journal of Zoology 43, 105-118. Chapter 9:Conclusions and Implications 232

Wells, M. C. (1978). Coyote senses in predation: environmental influences on their relative use. Behavioural Processes 3, 149-158.

Wilson, S. R., Carmack, M., Novotny, M., Jorgenson, J. W. and Whitten, W. K. (1978). delta3-Isopentenyl Methyl Sulfide. A New Terpenoid in the Scent Mark of the Red Fox (Vulpes vulpes). Journal of Organic Chemistry 43, 4675-4676.

Young, J. M., Friedman, C., Williams, E. M., Ross, J. A., Tonnes-Priddy, L. and Trask, B. J. (2002). Different evolutionary processes shaped the mouse and human olfactory receptor gene families. Human Molecular Genetics 11, 535-546.

Young, J. M. and Trask, B. J. (2002). The sense of smell: genomics of vertebrate odorant receptors. Human Molecular Genetics 11, 1153-1160. Appendix 1: A consideration of odour 233

Appendix 1:

A consideration of odour: how do mammals utilise their olfactory capabilities?

In Chapter 8, I investigated how tiger quolls Dasyurus maculatus respond to prey species odours. Although they readily followed scent trails back to their source, without the scent trails the quolls appeared unable to locate the odour source unless their nose was within approximately 30 cm of it. A review of the literature revealed a number of other studies where mammalian predators had a similarly low range of olfactory detection. However, there have been several anecdotal reports detailing carnivores locating prey by their odour over a distance of greater than a kilometre (Mech 1970;

Mills 1978) and a number of studies assume that mammalian predators are attracted to areas with an accumulation of prey species scent marks due to their odour (Banks et al.

2000; Koivula and Korpimäki 2001; Banks et al. 2002; Wolff 2004). In this appendix, I review what is known of olfaction as a distance receptor in mammals.

A1.1 Olfactory perception

In the mammalian olfactory system, odours are perceived when volatile molecules interact with odorant receptors in the fine cilia of the olfactory sensory neurons (Menini et al. 2004). These olfactory receptor cells are located in the nasal mucosa within the olfactory epithelium (Stockhorst and Pietrowsky 2004). In the nose of the domestic dog

Canis lupus familiaris, the olfactory epithelium covers an area of ~150 cm2 (Stoddart

1980). In comparison, in humans the olfactory epithelium covers an area of just 2-4 cm2

(Stoddart 1980) and it is generally assumed that the capacity of the human sense of Appendix 1: A consideration of odour 234

smell is inferior to that of most other mammal species (Doty 1985; Stockhorst and

Pietrowsky 2004). While it is true that dogs are more sensitive to odours than humans, insofar as they can detect chemicals at much lower concentrations (Marshall et al.

1981; Marshall and Moulton 1981), how much greater the distance over which a dog or any other mammal can actually detect an odour source remains largely unknown.

A1.2 Odour plume dynamics

In order to be detected, odour molecules above a certain threshold concentration must come into contact with the nasal epithelium (Stoddart 1976). Wright (1958) and Bossert and Wilson (1963) derived a series of mathematical formulae describing the diffusion of odours from a source, which allowed for the calculation of the distance from the odour source that the concentration of the odour would stay above this threshold of detection. These equations were based on; the strength of the source, the density of the volatiles in the odour, the rate of diffusion of these volatiles, the mean wind velocity, the horizontal and vertical profile of the wind, the gust profile of the wind, and the roughness of the ground. In still air, odours diffuse at a slow but equal rate in all directions away from the source, and the threshold of detection has a very limited range. Wind may increase the range of this threshold concentration by hundreds of metres or more, depending on its strength and the strength of the odour, but the increased ellipsoid of detection only occurs downwind from the odour source and the range of the threshold concentration upwind is actually decreased (Bossert and Wilson

1963)

The extent of such odour plumes above the threshold of detection have not been examined for mammals, however the matter has been extensively studied when it Appendix 1: A consideration of odour 235

comes to pheromone plumes and insects. It is recognised that time-averaged models such as those used by Wright (1958) and Bossert and Wilson (1963), and other

Gaussian models of odour plume dynamics developed subsequently (Elkinton et al.

1984; Stanley et al. 1985), do not take into account the complexity of the fluctuations within an odour plume caused by turbulent diffusion of the wind (Wright 1958; Aylor and Parlange 1976; Elkinton et al. 1984). Within an odour plume, there are both large and small scale fluctuations in odour concentration. Large scale fluctuations are the meanderings of the actual odour plume caused by the constant changes in direction of the wind (David et al. 1982; David et al. 1983; Elkinton et al. 1987). Even small changes in wind direction can result in the typical meanderings one can observe in any billow of smoke. Small scale fluctuations occur within the actual odour plume. The concentration of odour within a plume is highly variable, consisting of instantaneous peaks and intermittency (periods of zero concentration of odour) (Murlis and Jones

1981; Murlis et al. 2000), however the values of both the mean and peak odour concentration decreases with increasing distance from the odour source, and intermittency increases (Murlis et al. 2000). Habitat also has an effect on these characteristics. In comparisons between open fields and forests, odour plumes in forests have been found to have much greater variability in trajectory than in open fields

(Elkinton et al. 1987), and the duration of peak concentration and intermittency are both increased in the forest (Murlis et al. 2000).

A1.3 Location of odour sources by insects

Because of this high variability, insects do not simply follow a gradient of odour concentration as once thought, but rather fly directly into the wind (David et al. 1982;

David et al. 1983). Although this may cause the insect to exit the odour plume, upon Appendix 1: A consideration of odour 236

doing so they takes up the tactic of casting; a reiterative zigzag that progressively widens across the direction of the odourless wind until contact with the plume is re- established (David et al. 1983). As the insect approaches the odour source, the concentration increases and the intermittency decreases (Murlis et al. 2000), but the near location of the odour source is through a combination of both olfactory and visual stimuli (Farkas et al. 1974; Charlton and Carde 1990).

Despite such behavioural responses to plume variation, insects still often fail to locate an odour source. Elkinton et al. (1987) found that at a distance of 20 m, only 45% of moths which responded to an odour by wing-fanning actually arrived at the source.

When the distance increased to 40 m, this value dropped to 27%, and further to 17% and 8% at distances of 80 and 120 m respectively. Detection rates may be even lower for natural odour sources, as the synthetic pheromones used in these experiments were released at a much higher concentration than are actually released by the female insects in the wild (Charlton and Carde 1982; Elkinton et al. 1987).

A1.4 Location of odour sources by mammals

Although the distances over which mammals can detect and locate an odour source have not been tested empirically, several species have been observed to detect an odour source from over a kilometre away, when the wind and odour were strong enough.

Mills (1978) described a brown hyena Hyaena brunnea sniffing upwind and then changing direction into the wind, eventually finding the carrion source 1.9 km away.

Hyenas were said to sniff the air more often when the wind was blowing, and always in the direction of the wind (Mills 1978). The habit of brown hyenas apparently picking up the scent, moving quickly forward, only to revert to their original course without Appendix 1: A consideration of odour 237

having found anything may not have been understood by Mills (1978), but sounds like the behaviour of those moths which were unable to successfully track the odour plume back to its source. Mills (1978) also points out that when upwind of carrion, brown hyenas were unable to detect it. On one occasion Mech (1970) observed a pack of wolves Canis lupus detect three moose Alces alces from a distance of 1.5 miles (2.4 km), but he noted that the normal range of olfactory detection was within 300 yards

(274 m). In order to detect the moose, the wolves had to be directly downwind and upon detecting the scent headed directly upwind; if the prey was not upwind they remained undetected (Mech 1970). Similarly in 80 m x 80 m outdoor enclosures Wells

(1978) found that coyotes Canis latrans only detected rabbit carcasses from downwind, and detected them at greater distances as the wind velocity increased up to a distance of

10 m at a wind velocity of 45 km/h. Although this is several orders of magnitude less than 2.4 km, it was an outlier among their data; the average distance of detection for this wind speed was around only 5.5 m.

So, with a strong wind, mammalian predators are able to detect a strong odour from a stationary target such as a carcass over a longer distance, as long as they are downwind of the source, and even then they do not always locate the odour source. Odour plume dynamics are the same regardless of the animals detecting the odour, so presumably the predators travel upwind as long as they continue to receive occasional bursts of peak intensity of concentration. Should their upwind travel take them out of the plume then they may cast about trying to re-establish contact, but that may or may not be successful. This is also likely to be affected by their habitat. When the brown hyena located the gemsbok carcass over a 1.9 km distance it was in a large extremely open savanna (Mills 1978). Similarly, the coyotes located the rabbit carcasses over a distance Appendix 1: A consideration of odour 238

of up to 10 m in an open 80 m x 80 m outdoor enclosure. It is likely that location of an odour source would be more difficult in a closed environment such as a forest due to the greater variability in trajectory of the odour plume (Elkinton et al. 1987), and the increased rate of intermittency of the odour signal (Murlis and Jones 1981).

In still air, coyotes and hyenas can detect a scent from any direction, but over a much reduced distance, and take some time after initially detecting the odour, before locating the source (Mills 1978; Wells and Lehner 1978). Although Mills (1978) did not state what these reduced distances were for hyenas, for coyotes detection distances at wind speeds below 10 km/h declined to less than 1.5 m (Wells 1978). Österholm (1964) found that in a 10 x 10 m enclosure red foxes Vulpes vulpes could only use olfaction to detect pieces of meat placed on the ground from a distance of less than 1.5 m away.

Similar to this result, in the field domestic dogs were only able to detect incubating red grouse Lagopus lagopus from a distance of less than 0.5 m (Hudson 1992), whilst capercaillie Tetrao urogallus nests were only detected at distances closer than 1.5 m and black grouse T. tetrix nests were only detected at distances closer than 1.1 m

(Storaas et al. 1999).

These smaller distances of detection than many would assume does not mean that olfaction is less important than has previously been surmised. Among small mammals, olfaction has been found to be the more important than vision or audition in the near location of mealworms by wongai ningaui Ningaui ridei and grey-bellied dunnarts

Sminthopsis griseoventer (Huang 1986), as important as vision in the near location of dead crickets by grasshoppper mice Onychomoys leucogaster (Langley 1983), and golden hamsters Mesocricetus auratus (Langley 1985) and more important than vision Appendix 1: A consideration of odour 239

in spiny mice Acomys cahirinus (Langley 1988). But all of these experiments were conducted in cages with their longest basal length being less than 1.2 m. Although some terrestrial vertebrate prey may have a stronger odour than crickets and mealworms, the observations for larger predators mentioned above would seem to suggest that in still conditions 1.5 m is the average maximum distance of detection.

A1.5 Odour discrimination

However, distance is only one aspect of olfaction. The domestic dog’s greater olfactory capacity lies not just in their ability to detect low concentrations of odour; it is also their ability to recognise and discriminate between different odours which make them far superior olfactory detectors to humans (Adrian 1953; Moulton et al. 1960; Ewer 1973).

Dogs are able to recognise individual humans purely by scent (Brisbin and Austad

1991; Settle et al. 1994) even to the point of being able to discriminate between identical twins, if presented with the odours of both twins (Kalmus 1955; Hepper

1988). It is because of this ability to recognise and differentiate between a vast range of odours that has led to them being used by humans as chemical detectors in an ever increasing number of situations (Furton and Myers 2001; Lorenzo et al. 2003).

Dogs, like many other mammalian species, use these abilities to deduce information from conspecific scent marks about territory, identity, sex, and reproductive state

(Gorman and Trowbridge 1989). Many scent marks are extremely rich in lipids, which slows down the release of volatiles, increasing the longevity of the scent (Regnier and

Goodwin 1976; Brahmachary 1986). But if fewer volatiles are being released then subsequently the distance of detection will also be less. However, it is the longevity and amount of information they contain rather their distance of detection, which are the Appendix 1: A consideration of odour 240

most important aspects of scent marks. Scent marks are not placed at random, but rather in conspicuous often elevated positions (Kleiman 1966; Gorman 1980; Macdonald

1980). Animals can find these scat marks because they know where to look for them, much like human scatologists. They investigate areas where scent marks are likely to be deposited (Gorman and Trowbridge 1989), but cannot receive the information they supply until in close proximity. This link between visual conspicuousness and the ability to find scent marks has been clearly observed in dingoes C. lupus dingo. When urine and anal gland secretions were placed on the ground in an area devoid of conspicuous objects dingoes would walk over them without detecting them unless they happened to walk directly downwind (Corbett 1995). Similarly dingoes started to sniff at a previously ignored patch of bare ground, despite the lack of an odour source, when rakings similar to those made by dingoes after scent marking were created, and also sniffed at a grass tussock after it was planted in another part of the ignored bare patch

(Corbett 1995).

A1.6 Olfactory cues and predation

Mammalian predators are likely not only searching these conspicuous areas for conspecifics’ scent marks, but also for those of prey species. Weasels Mustela nivalis are known to be attracted to the urine of prairie deermice (Cushing 1984) and the urine of bank voles Clethrionomys glareolus and field voles Microtus agrestis (Ylönen et al.

2003), and tiger quolls are attracted to the urine and faeces of bush rats Rattus fuscipes and brown antechinus Antechinus stuartii (Chapter 8). The “search strategy” was defined by Smith (1974) as a basic set of rules of scanning and locomotion which results in the effective encountering of a specific distribution of food. Encountering the scent marks of prey species indicates such a specific distribution of food, causing the Appendix 1: A consideration of odour 241

predator to switch to a particular “search tactic”. Smith (1974) defined the “search tactic” as an adaptive change to scanning or locomotion occurring once a predator has arrived in a specific area where prey are available. One of the simplest forms of such a

“search tactic” is “area-restricted searching” (Tinbergen et al. 1967), but information deduced from scent marks about the prey species may also lead to a more specialised

“search tactic”. Banks et al. (2000) successfully predicted that voles with low mobility were more susceptible to predation by mustelids, as their proximity to accumulations of scent marks left them more susceptible to these predators’ search tactics. Prey may be sensitive to such risks, as house mice Mus domesticus change their behaviour towards scent marks in response to ferret Mustela putorius furo urinary odour, with dominant high-marking individuals significantly reducing their level of scent marking (Roberts et al. 2001).

The exploitation of scent trails from integumentary odours or deliberate scent marks left by prey may be one means by which many mammalian predators can overcome the difficulties associated with pursuit of airborne odour plumes. For example, Leyhausen

(1979) quotes Eibel-Eibesfeldt (1950) as stating that cats “seem to nose their way along the pungent “mouse tracks” marked with urine that mice make moving to and from their holes”. Tiger quolls can reportedly pick up the cross-trail of a rabbit (Fleay 1932), and will follow trails of bush rat and brown antechinus urine (Chapter 8). Mills (1978) described a brown hyena travelling 1.5 km with her nose to the ground, following the scent trail of a springbok carcasses that presumably had been dragged there by another predator. Spotted Hyenas Crocuta crocuta are able to follow scent trails, created by dragging carcass parts, three days later (Bearder 1977), however wolves on Isle Royale followed only very fresh moose tracks, and sometimes did not detect moose tracks only Appendix 1: A consideration of odour 242

minutes after the moose had created them (Mech 1970). This is probably due to the different nature of the two trails; the dragging of carcass parts is likely to leave behind blood, skin, hair, and other remains as longer term odour sources, whilst moose tracks are based solely on the odour imprint from moose hooves. Trained domestic dogs are known to be able to follow a trail left by humans up to a period of at least one hour later based on the odour of their footprints (Kalmus 1955; Mackenzie and Schultz 1987;

Steen and Wilsson 1990; Thesen et al. 1993; Wells and Hepper 2003), but in order to detect this odour hold their nose within 1 cm of the footprints (Thesen et al. 1993).

Similarly, American martens Martes americana are known to closely sniff the tracks of prey species and have been reported following them for distances over 600 m (Spencer and Zielinski 1983).

As reviewed in chapter 1, many prey use similar eavesdropping tactics to detect their predators as a means to assess their predation risk. But in the absence of wind, prey species are unlikely to be able to detect predator odours over a long distance. Laing

(1975) found that olfactory thresholds of laboratory brown rats Rattus norvegicus were the same as for humans. Dielenberg and McGregor (2001) found that laboratory rats didn’t respond to the odour of a cat collar until they were within 5cm, although

Blanchard et al (2003) claimed their rats responded to cat fur odour at a distance of

30cm. And house mice can only detect conspecific scent marks over a distance of a few centimetres (Hurst 2005).

Thus, in order to respond to a predator’s scent mark, prey species may actually have to come into close proximity with the scent mark to receive the signal. This implies a strategy of “detection-aversion” is needed if prey species are to use predator odours to Appendix 1: A consideration of odour 243

reduce their risks. Upon initially perceiving the predator odour their first impulse may be to further investigate before abandoning the area. For example, several deer species investigate the scent marks of tigers Panthera tigris at close range (Muller-Schwarze

1972). Meadow voles M. pennsylvanicus also spent more time sniffing weasel urine than the alternate guinea pig urine or the odourless control, but spent less time in total in the chamber where weasel odour was presented (Parsons and Bondrup-Nielsen

1996). Such detection-aversion behaviour might not be detected by remote measures of prey behaviour such as trapping which only measures the animal’s initial response to the stimuli. This could be why meadow voles in the same study did not avoid predator scented traps in the field (Parsons and Bondrup-Nielsen 1996). Similarly brown antechinus and long-nosed bandicoots Perameles nasuta showed no apparent avoidance of quoll-scented traps in the field (Chapters 3 and 4) but spent less time in the area where quoll scats were presented in a captive situation (Chapters 5 and 6).

A1.7 Conclusion

Odour detection is highly dependent on both wind speed and habitat. In low wind conditions and/or more closed environments such as forests, distances of detection are greatly reduced. If both predators and prey detect these odours over much smaller distances than is commonly assumed, then there are many implications. The likelihood that prey do not respond to predator odour like a barrier at some distance from the source, and that predators do not use prey species odour like some sort of radar, but rather that the perception of both take place at close range to the source, needs to be considered in the context of the results of experiments dealing with odour based predator/prey interactions. This lesser range of olfactory detection than is commonly assumed may also have serious implications for other aspects of mammalian ecology. Appendix 1: A consideration of odour 244

Bearing this possibility in mind may lead to a much greater understanding of the way in which odours are utilised by mammalian species.

A1.8 References

Adrian, E. D. (1953). The mechanism of olfactory stimulation in the mammal. Advancement of Science, London 9, 417-427.

Aylor, D. E. and Parlange, J.-Y. (1976). Turbulent Dispersion of Disparlure in the Forest and Male Gypsy Moth Response. Environmental Entomology 5, 1026- 1032.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2000). Nonlinearity in the predation risk of prey mobility. Proceedings of the Royal Society of London: Series B, Biological Sciences. 267, 1621-1625.

Banks, P. B., Norrdahl, K. and Korpimäki, E. (2002). Mobility decisions and the predation risks of reintroduction. Biological Conservation 103, 133-138.

Bearder, S. K. (1977). Feeding habits of spotted hyaenas in a woodland habitat. East African Wildlife Journal 15, 263-280.

Blanchard, D. C., Markham, C., Yang, M., Hubbard, D., Madarang, E. and Blanchard, R. J. (2003). Failure to Produce Conditioning With Low-Dose Trimethylthiazoline or Cat Feces as Unconditioned Stimuli. Behavioral Neuroscience 117, 360-368.

Bossert, W. H. and Wilson, E. O. (1963). The Analysis of Olfactory Communication Among Animals. Journal of Theoretical Biology 5, 443-469.

Brahmachary, R. L. (1986). Ecology and chemistry of mammalian pheromones. Endeavour 10, 65-68.

Brisbin, I. L. and Austad, S. N. (1991). Testing the individual odour theory of canine olfaction. Animal Behaviour 42, 63-69.

Charlton, R. E. and Carde, R. T. (1982). Rate and diel periodicity of pheromone emission from female gypsy moths, (Lymantria dispar) determined with a glass- absorption collection system. Journal of Insect Physiology 28, 423-430.

Charlton, R. E. and Carde, R. T. (1990). Orientation of Male Gypsy Moths, Lymantria Dispar (L.), to Pheromone Sources: The Role of Olfactory and Visual Cues. Journal of Insect Behavior 3, 443-469.

Corbett, L. (1995). The Dingo in Australia and Asia. UNSW Press, Sydney.

Cushing, B. S. (1984). A Selective Preference by Least Weasels for Oestrous Versus Dioestrous Urine of Prairie Deer Mice. Animal Behaviour 32, 1263-1265. Appendix 1: A consideration of odour 245

David, C. T., Kennedy, J. S. and Ludlow, A. R. (1983). Finding of a sex pheromone source by gypsy moths released in the field. Nature 303, 804-806.

David, C. T., Kennedy, J. S., Ludlow, A. R., Perry, J. N. and Wall, C. (1982). A reappraisal of insect flight towards a distant point source of wind-borne odor. Journal of Chemical Ecology 8, 1207-1215.

Dielenberg, R. A. and McGregor, I. S. (2001). Defensive behavior in rats towards predatory odors: a review. Neuroscience and Biobehavioral Reviews 25, 597- 609.

Doty, R. L. (1985). The primates III: humans. In Social odours in mammals. (edited by R. E. Brown and D. W. Macdonald). 2 pp. 804-832. Clarendon Press, Oxford.

Elkinton, J. S., Carde, R. T. and Mason, C. J. (1984). Evaluation of time-average dispersion models for estimating pheromone concentration in a deciduous forest. Journal of Chemical Ecology 10, 1081-1108.

Elkinton, J. S., Schal, C., Ono, T. and Carde, R. T. (1987). Pheromone puff trajectory and upwind flight of male gypsy moths in a forest. Physiological Entomology 12, 399-406.

Ewer, R. F. (1973). The Carnivores. Weidenfeld and Nicolson, London.

Farkas, S. R., Shorey, H. H. and Gaston, L. K. (1974). Sex Pheromones of Lepidoptera. Influences of Pheromone Concentration and Visual Cues on Aerial Odor-Trail Following by Males of Pectinophora gossypiella. Annals of the Entomological Society of America 67, 633-638.

Fleay, D. (1932). The rare dasyures (native cats). Victorian Naturalist 49, 63-69.

Furton, K. G. and Myers, L. J. (2001). The scientific foundation and efficacy of the use of canines as chemical detectors for explosives. Talanta 54, 487-500.

Gorman, M. L. (1980). Sweaty Mongooses and Other Smelly Carnivores. Symposium of the Zoological Society of London 45, 87-105.

Gorman, M. L. and Trowbridge, B. J. (1989). The role of Odor in the Social Lives of Carnivores. In Carnivore Behavior, Ecology, and Evolution. (edited by J. L. Gittleman) pp. 57-88. Cornell University Press, New York.

Hepper, P. G. (1988). The discrimination of human odour by the dog. Perception 17, 549-554.

Huang, C. (1986). Detection of prey by captive Ningaui ridei and Sminthopsis griseoventer (Marsupialia: Dasyuridae). Australian Mammalogy 10, 23-26.

Hudson, P. J. (1992). Grouse in Space and Time: The Population Biology of a Managed Gamebird. Game Conservancy Limited, Fordingbridge. Appendix 1: A consideration of odour 246

Hurst, J. L. (2005). Scent marking and social communication. In Animal Communication Networks. (edited by P. K. McGregor) pp. 219-243. Cambridge University Press, Cambridge.

Kalmus, H. (1955). The discrimination by the nose of the dog of individual human odours and in particular of the odours of twins. British Journal of Animal Behaviour 3, 25-31.

Kleiman, D. (1966). Scent Marking in the Canidae. Symposium of the Zoological Society of London 18, 167-177.

Koivula, M. and Korpimäki, E. (2001). Do scent marks increase predation risk of microtine rodents? Oikos 95, 275-281.

Laing, D. G. (1975). A comparative study of the olfactory sensitivity of humans and rats. Chemical Senses and Flavor 1, 257-269.

Langley, W. M. (1983). Relative importance of the distance senses in grasshopper mouse predatory behaviour. Animal Behaviour 31, 199-205.

Langley, W. M. (1985). Relative importance of distance senses in hamster predatory behavior. Behavioural Processes 10, 229-239.

Langley, W. M. (1988). Spiny Mouse's (Acomys cahirinus) use of its distance senses in prey localization. Behavioural Processes 16, 67-73.

Leyhausen, P. (1979). Cat behaviour: the predatory and social behaviour of domestic and wild cats. Garland STPM Press, New York.

Lorenzo, N., Wan, T., Harper, R. J., Hsu, Y.-L., Chow, M., Rose, S. and Furton, K. G. (2003). Laboratory and field experiments used to identify Canis lupus var familiaris active odor signature chemicals from drugs, explosives, and humans. Analytical and Bioanalytical Chemistry 376, 1212-1224.

Macdonald, D. W. (1980). Patterns of Scent Marking with Urine and Faeces Amongst Carnivores Communities. Symposium of the Zoological Society of London 45, 107-139.

Mackenzie, S. A. and Schultz, J. A. (1987). Frequency of Back-Tracking in the Tracking Dog. Applied Animal Behaviour Science 17, 353-359.

Marshall, D. A., Blumer, L. and Moulton, D. G. (1981). Odor detection curves for n- pentanoic acid in dogs and humans*. Chemical Senses 6, 445-453.

Marshall, D. A. and Moulton, D. G. (1981). Olfactory sensitivity to alpha-ionone in humans and dogs. Chemical Senses 6, 53-61.

Mech, L. D. (1970). The Wolf: the ecology and behavior of an endangered species. Natural History Press, New York.

Menini, A., Lagostena, L. and Boccaccio, A. (2004). Olfaction: From odorant molecules to the Olfactory Cortex. News in Physiological Science 19, 101-104. Appendix 1: A consideration of odour 247

Mills, M. G. L. (1978). Foraging Behaviour of the Brown Hyaena (Hyaena brunnea Thunberg, 1820) in the Southern Kalahari. Zeitshrift fuer Tierpsychologie 48, 113-141.

Moulton, D. G., Ashton, E. H. and Eayrs, J. T. (1960). Studies in olfactory acuity. 4) Relative detectability of n-aliphatic acids by the dog. Animal Behaviour 8, 117- 128.

Muller-Schwarze, D. (1972). Responses of young black-tailed deer to predator odors. Journal of Mammalogy 53, 393-394.

Murlis, J. and Jones, C. D. (1981). Fine-scale structure of odour plumes in relation to insect orientation to distant pheromone and other attractant sources. Physiological Entomology 6, 71-86.

Murlis, J., Willis, M. A. and Carde, R. T. (2000). Spatial and temporal structures of pheromone plumes in fields and forests. Physiological Entomology 25, 211-222.

Österholm, H. (1964). The significance of distance receptors in the feeding behaviour of the fox (Vulpes vulpes L.). Acta Zoologica Fennica 106, 1-31.

Parsons, G. J. and Bondrup-Nielsen, S. (1996). Experimental analysis of behaviour of meadow voles (Microtus pennsylvanicus) to odours of the short-tailed weasel (Mustela erminea). Ecoscience 3, 63-69.

Regnier, F. E. and Goodwin, M. (1976). On the Chemical and Environmental Modulation of Pheromone Release from Vertebrate Scent Marks. In Chemical Signals in Vertebrates. (edited by D. Muller-Schwarze and M. M. Mozell) pp. 115-134. Plenum Press, New York.

Roberts, S. C., Gosling, L. M., Thornton, E. A. and McClung, J. (2001). Scent-marking by male mice under the risk of predation. Behavioral Ecology 12, 698-705.

Settle, R. H., Sommerville, B. A., McCormick, J. and Broom, D. M. (1994). Human scent matching using specially trained dogs. Animal Behaviour 48, 1443-1448.

Smith, J. N. M. (1974). The food searching behaviour of two European Thrushes. II: The adaptiveness of the search patterns. Behaviour 49, 1-61.

Spencer, W. D. and Zielinski, W. J. (1983). Predatory Behavior of Pine Martens. Journal of Mammalogy 64, 715-717.

Stanley, B. H., Hummel, H. E. and Ruesink, W. G. (1985). Estimating maximum horizontal area of pheromone plumes. Journal of Chemical Ecology 11, 1129- 1146.

Steen, J. B. and Wilsson, E. (1990). How do dogs determine the direction of tracks? Acta Physiologica Scandinavica 139, 531-534.

Stockhorst, U. and Pietrowsky, R. (2004). Olfactory perception, communication, and the nose-to-brain pathway. Physiology and Behavior 83, 3-11. Appendix 1: A consideration of odour 248

Stoddart, D. M. (1976). Mammalian Odours and Pheromones. Edward Arnold Ltd, London.

Stoddart, D. M. (1980). Vertebrate olfaction. Endeavour 5, 9-14.

Storaas, T., Kastdalen, L. and Wegge, P. (1999). Detection of forest grouse by mammalian predators: A possible explanation for high brood losses in fragmented landscapes. Wildlife Biology 5, 187-192.

Thesen, A., Steen, J. B. and Doving, K. B. (1993). Behaviour of dogs during olfactory tracking. Journal of experimental Biology 180, 247-251.

Tinbergen, N., Impekoven, M. and Franck, D. (1967). An experiment on spacing-out as a defence against predation. Behaviour 28, 307-321.

Wells, D. L. and Hepper, P. G. (2003). Directional tracking in the domestic dog, Canis familiaris. Applied Animal Behaviour Science 84, 297-305.

Wells, M. C. (1978). Coyote senses in predation: environmental influences on their relative use. Behavioural Processes 3, 149-158.

Wells, M. C. and Lehner, P. N. (1978). The relative importance of the distance senses in coyote predatory behaviour. Animal Behaviour 26, 251-258.

Wolff, J. O. (2004). Scent marking by voles in response to predation risk: a field- laboratory validation. Behavioral Ecology 15, 286-289.

Wright, R. H. (1958). The Olfactory Guidance of Flying Insects. Canadian Entomologist 90, 81-89.

Ylönen, H., Sundell, J., Tiilikainen, R., Eccard, J. A. and Horne, T. (2003). Weasels' (Mustela nivalis nivalis) preference for olfactory cues of the vole (Clethrionomys glareolus). Ecology 84, 1447-1452.