Phenotypic Evolution in the Invasive Cane Toad (Rhinella Marina): Adaptations for Dispersal

Home , Rhinella, Rhinella arenarum, Rhinella icterica

Phenotypic evolution in the invasive cane toad (Rhinella marina): adaptations for dispersal

A thesis submitted in fulfillment of the requirements for the degree of

Doctor of Philosophy

Cameron Marshall Hudson

School of Life and Environmental Sciences

The University of Sydney

Australia

January 2017 Supervisor Statement

This is to certify that I have been Cameron’s primary supervisor during his PhD candidature and have overseen the production of this thesis. In my opinion Cameron’s thesis is sufficiently well presented to be examined, and it does not exceed the prescribed word limit or any extended word limit for which prior approval has been granted.

Rick Shine

University of Sydney

Declaration

I hereby certify that the work presented in this thesis is my own (except where I provide references specifically acknowledging otherwise). This thesis has not been submitted to attain the degree of Doctor of Philosophy at another University.

Cameron Hudson

Sydney, January 2017

Preface

In accordance with the guidelines for a PhD thesis in the Faculty of Science at the University of Sydney, all data chapters of this thesis are presented as stand-alone manuscripts that are either published in peer-reviewed journals, currently under consideration for publication, or in preparation for submission. This method of presentation means that there is some unavoidable repetition of background information and methodology in each chapter.

All of the chapters within this thesis have been written with various co-authors, including my supervisors, Professor Richard Shine and Dr. Gregory Brown, along with other colleagues: Dr. Ben Phillips, Dr. Colin McHenry, Matthew McCurry, and Petra Lundgren.

These individuals contributed to the initial planning of the research, participated in discussions and final editing of the manuscripts, and in some cases aided in laboratory experiments. However, I conducted all of the research associated with this thesis, including the field work, laboratory experiments, data analysis, manuscript drafting and writing.

Approval for this research was provided by the University of Sydney Animal Care and Ethics Committee (Approval number: AEC 6075).

iii

Acknowledgements

I owe thanks to a great number of people for their help, support, friendship, and encouragement during these past 3.5 years, and I wish to expound their virtues at length.

After all, it’s not every day that you get a chance to express your gratitude to the people that you care about in a printed volume. I can only hope that the words I’ve chosen here will convey how sincerely thankful I am to all of you.

I must begin with my supervisor, Professor Rick Shine, for without him, none of this would have been possible. Rick, I’m sure that by now, after supervising so many students, you’ve heard every variation on how to sing your praises. That, in itself is a testament to how inspiring and outstanding you are as a scientist. I consider myself incredibly lucky to have been a part of your lab, and I cannot thank you enough for the opportunities you have afforded me. Throughout this PhD I have both literally, and figuratively, travelled around the world in search of cane toads to measure. You have nurtured my toad-measuring obsession with enthusiasm, and consistently amaze me with your intellect. I can say with all honesty that this PhD has been a dream come true. So thank you for taking a chance on an excitable

Canadian from the Queen’s University Biological Station.

To my co-supervisor and compatriot, Dr. Greg Brown, you have proven to be an endless font of knowledge and wisdom throughout my time here. I sometimes worry that I have come to rely on you too much, because you are so consistently helpful. I will miss popping into your office to pester you about toads and statistics. Thank you for sharing your little slice of the Territory with me. I hope that one day I can return to Middle Point to find you still driving the Fogg Dam wall in search of keelbacks.

My fellow Shine labbers, my Bufo Bandits, my Rhinella Rangers, it has been a wild ride. We’ve survived our various cohabitations in the Top End, Hawai’i, and French Guiana, and I think we’ve come out better for it. I hate to lump you all into a giant wall of bold text

iv here, but there are just too damn many of you! So thank you to Melanie Elphick (especially for all of her hard work making this thesis coherent), Dr. Michael Crossland (who deserves special mention for putting up with me and my various office pets for three years), Dr. Jayna

DeVore, Dr. Matt Greenlees, Dr. Camila Both, Dr. Simon Ducatez, Dr. Wei Chen, Dr.

Chris Friesen, Georgia Ward-Fear, Jodie Gruber, Sam McCann, Greg Clarke, Georgia

Kosmala, Uditha Wijethunga, Dan Natusch, Chalene Bezzina, Sarsha Gorissen, Dan

Selechnik, Damien Holden, Felicity Nelson, Patt Finnerty and Katarina Stuart. I’ve made some fantastic memories with you all, and will miss you dearly once I’ve left Australia.

Of course I need to give thanks to our favourite invasive anuran, Rhinella marina.

Despite how much people in Australia tend to hate them, I find it hard not to love the little toads. After all, it’s not their fault that they’re here, and in a bizarre way, their introduction to

Australia in the 1930s lead to all of my PhD experiences. So thank you to the 5000+ toads that I’ve measured over the course of my research for their unwilling participation in our study of evolution.

I’d like to thank my collaborators, Dr. Ben Phillips, Dr. Marta Vidal-García, Dr.

Colin McHenry, Matthew McCurry, Petra Lundgren, Dr. Bill Mautz, Michelle Quayle,

Crystal Kelehear and James Mertins for all of their time and hard work. It was truly a pleasure working with all of you, and I can only hope that we will continue to do so in the future.

During my time here I’ve been lucky enough to interact with some wonderful people who may or may not be involved in cane toad research, but have proven to be great friends along the way. Many thanks to Martin Mayer, George Madani, Maddie Sanders, Kevin

Donmoyer, Tiago Dalcin, Damian Lettoof, Dr. Cathy Shilton, Jack Reid, Andrea West,

Lee-Ann Rollins, and Mark Richardson. I also can’t leave out my favourite non-human

Australian, Trampers the frill-neck, one solid lizard.

My former academic supervisors, Dr. Stephen Lougheed, Dr. Jinzhong Fu, and Dr.

Kat Stewart deserve special thanks for laying the foundation of my academic career, and propelling me on this life trajectory. Without your help and mentoring during my undergraduate and master’s degrees, I do not think I would have made it to where I am today.

To the Hudson, Fleming, Simmons, Lemmen, Penfound, and Henson families, thank you all for being so amazing. I have always felt a strong sense of love and encouragement from you. I could not wish for a better group of people to call my family. I must thank my brother Dr. Zachary Hudson, who I often credit as my inspiration to become a scientist for his guidance over the years, and my parents Michael and Susan Hudson for everything. I can’t really think of a better way to put it into words. Thank you mom and dad, for everything.

Finally, to my darling Kimberley Lemmen, I know that I’ve asked a lot of you since moving to Australia. The distance alone would be too daunting for most people, not to mention the unique isolation of Middle Point, but you’ve stuck by my side throughout.

Through all the international flights, dropped calls, and nights apart you have patiently waited for me. At times I feel undeserving of your unwavering devotion to our relationship, but these past seven years with you have been truly wonderful. I cannot wait to see where life takes us next. And so, my love, I dedicate this thesis to you.

Table of Contents

Supervisor Statement ...... i

Declaration ...... ii

Preface...... iii

Acknowledgements ...... iv

Table of Contents ...... vii

Abstract ...... 1

Chapter 1. General Introduction and Thesis Synopsis ...... 3

Literature Cited ...... 8

Chapter 2. Virgins in the Vanguard: Low Reproductive Frequency in Invasion-Front

Cane Toads ...... 12

Abstract ...... 13

Introduction ...... 14

Materials and Methods ...... 15

Results ...... 18

Discussion ...... 21

Acknowledgements ...... 23

Literature Cited ...... 23

Chapter 3. Athletic Anurans: The Impact of Morphology, Ecology, and Evolution on

Climbing Ability in Invasive Cane Toads ...... 27

Abstract ...... 28

Introduction ...... 29

vii

Materials and Methods ...... 32

Results ...... 38

Discussion ...... 43

Acknowledgements ...... 47

Literature Cited ...... 47

Supporting Information ...... 51

Chapter 4. Constructing an Invasion Machine: The Rapid Evolution of a Dispersal-

Enhancing Phenotype During the Cane Toad Invasion of Australia ...... 52

Abstract ...... 53

Introduction ...... 54

Materials and Methods ...... 55

Results ...... 57

Discussion ...... 68

Acknowledgements ...... 72

Literature Cited ...... 72

Supporting Information ...... 78

Chapter 5. It’s Lonely at the Front: Contrasting Evolutionary Trajectories in Male and

Female Invaders ...... 87

Abstract ...... 88

Introduction ...... 89

Materials and Methods ...... 92

Results ...... 96

Discussion ...... 105

Acknowledgements ...... 109

viii

Literature Cited ...... 109

Chapter 6. Effects of Toe-Clipping on Growth, Body Condition, and Locomotion of

Cane Toads (Rhinella marina) ...... 117

Abstract ...... 118

Introduction ...... 119

Materials and Methods ...... 121

Results ...... 122

Discussion ...... 126

Acknowledgements ...... 128

Literature Cited ...... 128

Chapter 7. The Not-So-Great Escape: Evolutionary Shifts in Anti-predator Responses of Cane Toads ...... 133

Abstract ...... 134

Introduction ...... 135

Materials and Methods ...... 137

Results ...... 139

Discussion ...... 148

Acknowledgements ...... 152

Literature Cited ...... 152

Appendix A. First report of exotic ticks (Amblyomma rotundatum) parasitizing invasive cane toads (Rhinella marina) on the island of Hawai’i ...... 161

Abstract ...... 162

Introduction ...... 163

Materials and Methods ...... 164

Results ...... 165

Discussion ...... 168

Conclusions ...... 170

Acknowledgements ...... 172

Literature Cited ...... 172

Appendix B. Sex and weaponry: the distribution of toxin-storage glands on the bodies of male and female cane toads (Rhinella marina) ...... 177

Abstract ...... 178

Introduction ...... 179

Materials and Methods ...... 181

Results ...... 185

Discussion ...... 191

Acknowledgements ...... 196

Literature Cited ...... 196

Abstract

Invasive species provide a rare opportunity to study how organisms adapt when colonising novel environments. Despite the negative ecological impacts caused by the introduction of alien species, biological invasions act as natural experiments that we can exploit for ecological and evolutionary research. This is particularly true for introductions with precise geographic and historical records, such as the cane toad (Rhinella marina) introduction to

Australia. Arguably the most successful invasive anuran worldwide, the cane toad was first introduced to the Hawai’ian Islands in 1932, and subsequently to northeastern Queensland in

1935. Over the past 82 years, R. marina has spread rapidly and at an accelerating pace throughout northern and eastern Australia, causing massive ecological disturbances in its wake. This acceleration of dispersal capability is a well-documented phenomenon in invasive organisms that can occur via a combination of natural selection, and spatial sorting. Although the increasing rate of cane toad dispersal has been extensively researched, my thesis aims to address questions about the evolutionary changes that have taken place throughout the process. Namely, what morphological, behavioural, or physiological shifts in cane toad phenotypes are associated with increased dispersal ability?

I began by taking gross morphological measurements from toads across a transect through their northern Australian range that included long-colonised and invasion-front populations. Additional populations from a prior invasion (Hawai’i), and the native range

(French Guiana) were later added to compare the morphology of individuals from the source populations. I focused on the components of each limb (hand, radioulna, humerus [forelimb]; femur, tibiofibula, foot [hindlimb]), as limb morphology is strongly linked to locomotor ability, but also collected data on mass, snout-vent length, head width, and parotoid gland shape. A subset of Australian individuals were collected from the edges of the range, representing the oldest and newest toad populations to be used in a common-garden breeding

1 experiment. These individuals, and their resulting offspring were held in captivity and subjected to a series of performance trials. A separate group of individuals from invasion- front and range-core populations was used for Computerised X-ray Tomography (CT) scanning for precise geometric morphometric comparison of skeletal structure.

Captive breeding of cane toads from the invasion-front (in Western Australia) and long-colonised areas (in Queensland) allowed me to control for the influence of rearing environment on common-garden F1 individuals. This also enabled analyses on heritability of morphology and performance traits by comparing offspring to their parents, and siblings to each other. Common-garden F1 offspring were raised over a period of 27 months, and measured repeatedly throughout ontogeny. Performance trials consisted of climbing trials

(where toads would have to escape from a mesh tube by climbing vertically) and anti- predator raceway trials (where toads were encouraged to hop down a raceway by prodding).

These trials were conducted on wild and captive toads.

Within this thesis I document reproductive differences between invasion-front and long-colonised populations, significant geographic variation in locomotor performance, regional changes to skeletal structure, shifts in sexual dimorphism with time since colonisation, and heritability of behavioural and morphological traits; and I demonstrate that these changes have arisen via the rapid evolution of a high-dispersal phenotype during the invasion process.

Chapter 1

General Introduction

The impact of invasive species on native ecosystems has attracted extensive research, because introduced species can have devastating effects on the flora and fauna with which they interact (Williamson, 1996; Mack et al., 2000; Pimentel et al., 2000). Introductions can be deliberate (e.g. as a biological control; a conservation initiative; as a resource; for aesthetic reasons) or accidental (e.g. stowaways; escaped exotic pets: Lever, 2001). The resulting impact is highly dependent on the biology of the invading organism, its ability to colonise novel habitats, and the ecosystem’s ability to cope with the introduced species. Despite their potential for causing environmental catastrophes, in some cases the impacts of introduced species are unclear, negligible, or even positive (Gurevitch & Padilla, 2004; Wonham et al.,

2005; King et al., 2006). Extensive research is often required to understand complex ecological interactions.

Anthropogenic introductions provide a unique opportunity to study evolution on an ecological timescale as the introduced species and native biota adapt to each other’s presence.

As an invading population colonises novel habitats, it is likely to diverge from its source population via drift, spatial sorting, and natural selection. Thus, successful invasions constitute natural experiments that allow exceptional opportunities for ecological and evolutionary research (Carroll et al., 2007; Shine, 2012).

One common shift during biological invasions concerns the speed at which the invasion proceeds, with a general trend for an accelerated rate of invasion over time (Sharov

& Liebhold, 1998; Hastings et al., 2005; Chuang & Peterson, 2015). At least two different evolutionary processes might be responsible for such an acceleration. First, traits that enhance dispersal ability may also enhance individual fitness, for example if individuals at the

3 invasion front have access to more abundant resources (perhaps because of reduced conspecific densities and thus, lessened competition: Brown et al., 2013). Second, alleles that confer increased dispersal rates will tend to accumulate at an expanding range edge via

“spatial sorting”: that is, a tendency for fast-dispersing individuals to interbreed at the increasingly fast-spreading invasion vanguard, simply because slow-moving individuals are left behind (Shine et al., 2011). Both of these processes should result in phenotypic evolution, with the invasion front increasingly dominated by individuals with traits that enhance their rate of dispersal. If the invasion covers a broad geographic area, and enough time has passed, phenotypic differences should be identifiable between “new” and “old” populations. The invasion of cane toads (Rhinella marina; formerly Bufo marinus: Frost et al., 2006; Pramuk,

2006) through tropical Australia appears to have produced exactly these kinds of divergences

(Phillips et al., 2006), but detailed studies on this system have produced conflicting results

(Llewelyn et al., 2010; Tingley et al., 2012; Tracey et al., 2012). My PhD research aimed to resolve these conflicts by gathering a more extensive dataset – including measurements on a broader range of traits, and over a larger geographical area than has been the case with previous research.

The cane toad is a large and highly toxic anuran native to Central and South America.

It is arguably the most successful invasive anuran species worldwide (Lowe et al., 2000), having been introduced to more than 40 countries to date (Lever, 2001). Since their introduction to northern Queensland in 1935 as a biological control for the greyback and

Frenchi cane beetles (Dermolepida albohirtum and Lepidiota frenchi: Mungomery, 1936) the toads have rapidly expanded their range to occupy more than 1.2 million km2 of the

Australian landscape (Urban et al., 2007), and ultimately may occupy 2 million km2 (based on predictive models: Sutherst et al., 1996). The invasion has caused massive ecological disturbances with the primary impacts on native fauna being lethal toxic ingestion by

4 predators (Shine, 2010). The rate of toad invasion is increasing each year (10 km/annum in the 1960s to 55 km/annum in 2000: Phillips et al., 2007) with invasion-front toads exhibiting higher daily dispersal rates (Alford et al., 2009), greater directional yearly displacement

(Lindström, 2013; Brown et al., 2014) and greater relative leg lengths (RLL: Phillips et al.,

2006) than conspecifics from long-established populations in Queensland. Similarly, toads invading drier areas of inland Australia exhibit an altered skin permeability to water (Tingley

& Shine, 2011), and toads invading montane areas rapidly acclimate to lower temperatures

(McCann et al., 2014). These observations suggest that toads at the vanguard are adapting to the environments they invade, as well as evolving high-dispersal-rate phenotypes, thereby increasing the rate at which they can colonise new areas (Shine et al., 2011).

Because the current distribution of toads within Australia spans a variety of habitat types, from wet tropical rainforest to xeric scrubland, it is reasonable to expect that toads from geographically disparate populations should exhibit different phenotypes to cope with the environmental challenges imposed by the habitat they occupy. To further examine the phenotypic evolution of R. marina, I sampled toads spanning the invaded Australian range, along with three Hawai’ian islands (Hawai’i, Maui, and O’ahu) and a native range population in French Guiana, comparing the morphology and performance of toads from long- established populations to those from recently invaded areas. To determine the heritability of toad phenotypes I conducted a common-garden breeding experiment on Australian toads, comparing the morphology and performance of F1 individuals to that of their parents.

Rhinella marina provides exceptional opportunities for the study of phenotypic evolution because of precise historical records (e.g. dates and times of toad arrival at sites across the continent), a wide geographic range spanning various habitat types, extensive prior research, and a single origin in time and space.

My thesis is presented as a series of chapters representing stand-alone manuscripts that are either published in peer-reviewed journals, currently under consideration for publication, or in preparation for submission. This method of presentation means that there is some unavoidable repetition of background information and methodology throughout.

Chapter 2 documents a reduction in reproductive rate in highly dispersive invasion- front populations of cane toads in tropical Australia. If a high reproductive rate slows down an individual’s dispersal, vanguard individuals should exhibit lower reproductive output than conspecifics from long-colonised areas.

Chapter 3 examines geographic divergence in climbing ability in cane toads from four regions in Australia, plus two sites on the island of Hawai’i. The intention was to compare the morphology of invasive cane toad populations, and assess their performance through a physical challenge (i.e. climbing out of a pit). Although it is influenced by morphology, climbing ability in wild-caught cane toads appears to be driven primarily by local environmental conditions that facilitate and/or reward arboreal activity (i.e. foraging for arboreal prey; reaching water sources).

Chapter 4 presents three-dimensional morphometric analyses of toads from both range-core and invasion-front populations that I used to investigate the morphological changes that have accompanied the toads’ dramatic acceleration. Morphology of heads, limbs, pectoral girdles and pelvic girdles differed significantly between toads from the two areas, with invasion-front toads exhibiting wider forelimbs, narrower hindlimbs and more compact skulls. Those changes plausibly reflect an increased reliance on a bounding gait.

Chapter 5 explores the morphological divergence in limb length between Australian and Hawai’ian toad populations. I measured relative lengths of forelimbs and hindlimbs of

>3000 field-caught adult cane toads (Rhinella marina) from 67 sites in Hawai’i and Australia

(1 to 80 years post-colonisation), along with 489 captive-bred individuals from the common-

6 garden experiment. As cane toads spread from east to west across Australia, the ancestral condition (long limbs, especially in males) was modified. Limb dimensions showed significant heritability (2–17%), consistent with the possibility of an evolved response. Cane toad populations thus have undergone a major shift in sexual dimorphism in relative limb lengths during their brief (81-year) spread through tropical Australia.

Chapter 6 uses a long-term mark-recapture dataset collected by Dr. Greg Brown to investigate the impact of toe removal on body condition, locomotion, and growth rates for mass and snout-urostyle length of toads. This was done primarily to assess whether my toe- clipping method for marking captive toads would influence their locomotor ability or survival.

Toe-clipping had no significantly negative effects on cane toad growth, condition or locomotion.

Chapter 7 presents the results from field and laboratory trials on a toad’s response to simulated predation. By encouraging toads to hop along an artificially constructed raceway I collected data on locomotor parameters and behavioural/physiological responses. I found differences in anti-predator response between populations (e.g. in flight response and toxin exudation) as well as differences in performance (e.g. hop distance and speed).

Literature Cited

Alford, R.A., Brown, G.P., Schwarzkopf, L., Phillips, B.L., Shine, R. 2009. Comparisons

through time and space suggest rapid evolution of dispersal behaviour in an invasive

species. Wildl. Res. 36: 23–28.

Alford, R.A., Cohen, M.P., Crossland, M.R., Hearnden, M.N., James, D., Schwarzkopf, L.

1995. Population biology of Bufo marinus in northern Australia. Pages 173–181 in

Finlayson, M., ed., Wetland Research in the Wet-Dry Tropics of Australia. Office of the

Supervising Scientist Report 101, Canberra, Australia.

Brown, G.P., Kelehear, C., Shine, R. 2013. The early toad gets the worm: cane toads at an

invasion front benefit from higher prey availability. J. Anim. Ecol. 82: 854–862.

Brown, G.P., Phillips, B.L., Shine, R. 2014. The straight and narrow path: the evolution of

straight-line dispersal at a cane-toad invasion front. Proc. R. Soc. B. 281: 20141385.

Carroll, S.P., Hendry, A.H., Reznick, D., Fox, C.E. 2007. Evolution on ecological time scales.

Funct. Ecol. 21: 387–393.

Chuang, A., Peterson, C.R. 2015. Expanding population edges: theories, traits, and trade-offs.

Glob. Change Biol. 22: 494–512.

Frost, D.R., Grant, T., Faivovich, J., Bain, R.H., Haas, A., Haddad, C.F.B., De Sa´, R.O.,

Channing, A., Wilkinson, M., Donnellan, S.C., Raxworthy, C.J., Campbell, J.A., Blotto,

B.L., Moler, P., Drewes, R.C., Nussbaum, R.A., Lynch, J.D., Green, D.M., Wheeler,

W.C. 2006. The amphibian tree of life. Bull. Am. Mus. Nat. Hist. 297: 1–370.

Gurevitch, J., Padilla, D.K. 2004. Are invasive species a major cause of extinctions? Trends

Ecol. Evol. 19: 470–474.

Hastings, A., Cuddington, K., Davies, K.F., Dugaw, C.J., Elmendorf, S., Freestone, A.,

Harrison, S., Holland, M., Lambrinos, J., Malvadkar, U., Melbourne, B.A., Moore, K.,

Taylor, C., Thomson, D. 2005. The spatial spread of invasions: new developments in

theory and evidence. Ecol. Lett. 8: 91–101.

King, R.B., Ray, J.M., Stanford, K.M. 2006. Gorging on gobies: beneficial effects of alien

prey on a threatened vertebrate. Can. J. Zool. 84: 108–115.

Lever, C. 2001. The Cane Toad: The History and Ecology of a Successful Colonist. Westbury

Academic and Scientific Publishing, Otley, West Yorkshire, UK.

Lindström, T., Brown, G.P., Sisson, S.A., Phillips, B.L., Shine, R. 2013. Rapid shifts in

dispersal behavior on an expanding range edge. Proc. Natl. Acad. Sci. USA 110: 13452–

13456.

Llewelyn, J., Phillips, B.L., Alford, R.A., Schwarzkopf, L., Shine, R. 2010. Locomotor

performance in an invasive species: cane toads from the invasion front have greater

endurance, but not speed, compared to conspecifics from a long colonised area.

Oecologia 162: 343–348.

Lowe, S., Browne, M., Boudjelas, S., De Poorter, M. 2000. 100 of the World’s Most Invasive

Alien Species: A Selection from the Global Invasive Species Database. Invasive Species

Specialist Group, Auckland, New Zealand.

Mack, R.N., Simberloff, D., Lonsdale, W.M., Evans, H., Clout, M., Bazzaz, F. 2000. Biotic

invasions: causes, epidemiology, global consequences and control. Issues Ecol. 5: 1–20.

McCann, S., Greenlees, M.J., Newell, D., Shine R. 2014. Rapid acclimation to cold allows

the cane toad to invade montane areas within its Australian range. Funct. Ecol. 28:

1166–1174.

Mungomery, R.W. 1936. A survey of the feeding habits of the giant toad (Bufo marinus) and

notes on its progress since introduction into Queensland. Proc. Qld Soc. Sugar Cane

Technol. 1936: 63–74.

Phillips, B.L., Shine, R. 2005. The morphology, and hence impact, of an invasive species (the

cane toad, Bufo marinus): changes with time since colonisation. Anim. Conserv. 8: 407–

413.

Phillips, B.L., Brown, G.P., Webb, J.K., Shine, R. 2006. Invasion and the evolution of speed

in toads. Nature 439: 803.

Phillips, B.L., Brown, G.P., Greenlees, M., Webb, J.K., Shine, R. 2007. Rapid expansion of

the cane toad (Bufo marinus) invasion front in tropical Australia. Austral Ecol. 32: 169–

176.

Pimentel, D., Lach, L., Zuniga, R., Morrison, D. 2000. Environmental and economic costs of

nonindigenous species in the United States. Bioscience 50: 53–65.

Pramuk, J. 2006. Phylogeny of South American Bufo (Anura: Bufonidae) inferred from

combined evidence. Zool. J. Linn. Soc. 146: 407–452.

Sharov, A.A., Liebhold, A.M. 1998. Bioeconomics of managing the spread of exotic pest

species with barrier zones. Ecol. Appl. 8: 833–845.

Shine, R. 2010. The ecological impact of invasive cane toads (Bufo marinus) in Australia. Q.

Rev. Biol. 85: 253–291.

Shine, R., Brown, G.P., Phillips, B.L. 2011. An evolutionary process that assembles

phenotypes through space rather than through time. Proc. Natl. Acad. Sci. USA 108:

5708–5711.

Shine, R. 2012. Invasive species as drivers of evolutionary change: cane toads in tropical

Australia. Evol. Appl. 5: 107–116.

Sutherst, R.W., Floyd, R.B., Maywald, G.F. 1996. The potential geographical distribution of

the cane toad, Bufo marinus L. in Australia. Conserv. Biol. 10: 294–299.

Tingley, R., Shine, R. 2011. Desiccation risk drives the spatial ecology of an invasive anuran

(Rhinella marina) in the Australian semi-desert. PLoS One 6: e25979.

Tingley, R., Greenlees, M.J., Shine, R. 2012. Hydric balance and locomotor performance of

an anuran (Rhinella marina) invading the Australian arid zone. Oikos 121: 1959–1965.

Tracey, C.R., Christian, K.A., Baldwin, J., Phillips, B.L. 2012. Cane toads lack physiological

enhancements for dispersal at the invasive front in Northern Australia. Biol. Open 1: 37–

42.

Urban, M.C., Phillips, B.L., Skelly, D.K., Shine, R. 2007. The cane toad’s (Chaunus [Bufo]

marinus) increasing ability to invade Australia is revealed by a dynamically updated

range model. Proc. R. Soc. B 274: 1413–1419.

Williamson, M. 1996. Biological Invasions. Chapman and Hall, London.

Wonham, M.J., O’Connor, M., Harley, C.D.G. 2005. Positive effects of a dominant invader

on introduced and native mudflat species. Mar. Ecol. Prog. Ser. 289: 109–116.

Chapter 2

Virgins in the Vanguard: Low Reproductive Frequency in Invasion-front

Cane Toads

Manuscript published as:

Hudson, C.M., Phillips, B.L., Brown, G.P., Shine, R. 2015. Virgins in the vanguard: low reproductive frequency in invasion-front cane toads. Biol. J. Linn. Soc. 116: 743–747.

Abstract

The rapid evolution of increased dispersal rate during a population’s range expansion provides a unique opportunity to detect trade-offs between dispersal and reproduction. If a high reproductive rate slows down an individual’s dispersal, vanguard individuals should exhibit a lower reproductive output than conspecifics from long-colonised areas. In the present study, we demonstrate a reduction in reproductive rate in highly dispersive invasion- front populations of cane toads in tropical Australia.

Key-words: alien species, anuran, Bufo marinus, invader, life history, reproductive output, trade-off.

Introduction

On an advancing range edge, dispersal rapidly evolves upwards because of a conspiracy of evolutionary forces (Travis & Dytham, 2002; Phillips et al., 2010a; Perkins et al., 2013). First, there is ‘spatial sorting’ of dispersal phenotypes: only the most dispersive individuals reach the front each generation (Shine et al., 2011). Second, individuals on the invasion front may obtain a fitness advantage through reduced density of intraspecific competitors at the front

(Phillips et al., 2010a). Finally, increased relatedness may also lead to increased dispersal to avoid kin competition (Kubisch et al., 2013). Thus, spatial sorting, natural selection, and kin selection combine to drive the rapid evolution of dispersal in vanguard populations. These forces can generate large shifts. Cane toads on the invasion front in tropical Australia, for example, disperse more than five times faster than conspecifics in the range core (Alford et al., 2009). How is this possible, given the costs or trade-offs associated with faster dispersal?

Individual-based models suggest that evolutionary pressures on the invasion front can push dispersal upwards despite trade-offs with other fitness aspects (Belichon et al., 1996; Travis

& Dytham, 2002; Burton et al., 2010; Cote et al., 2010; Bonte et al., 2012). Nonetheless, such trade-offs ultimately limit the rate at which a population spreads (Burton et al., 2010).

Although implicit in many models of dispersal evolution, a trade-off between dispersal and other fitness components (Ronce, 2007) is difficult to demonstrate empirically

(Clobert et al., 2001; Bonte et al., 2012; Ronce & Clobert, 2012). With the exception of flight-fecundity trade-offs in insects (Karlsson & Johansson, 2008), there are few clear-cut cases of straightforward trade-offs between dispersal and other life-history aspects. More often, studies report complex syndromes that incorporate patterns of covariation among behavioural, physiological, and morphological traits reflecting the actions of phenotypic plasticity, as well as adaptation (Belichon et al., 1996; Cote et al., 2010; Ronce & Clobert,

2012). If invasion fronts push phenotypes towards higher dispersal rates, however, then

14 invasive species represent excellent systems for uncovering these trade-offs: high-dispersal phenotypes (at the invasion front) are spatially separated from lower-dispersal phenotypes

(other areas), and so we can simply measure reproductive rates from different areas (rather than having to measure both reproductive and dispersal rates for individuals in the same area).

Such spatial separation, however, introduces the confounding effects of local resource availability (which modifies reproductive rate); thus, we need to measure reproductive rates under standard (captive) conditions. In the course of our studies on invasive cane toads in tropical Australia (Lever, 2001; Shine, 2010), we have accumulated extensive data (although not originally gathered for that purpose) that can be used to compare reproductive output in individuals collected from geographically disparate populations but maintained under standard conditions. These data enable us to look for the predicted trade-off between dispersal and reproduction.

Materials and Methods

Cane toads (Rhinella marina, Linnaeus 1758) are large bufonid anurans from Central and

South America (Lever, 2001). Released in Australia in 1935, toads have spread at an accelerating pace (Urban et al., 2008). Compared to conspecifics from long-established populations, invasion-front toads move more often, further, and in straighter paths (Alford et al., 2009; Lindström et al., 2013; Brown et al., 2014), and also show modifications of physiology and morphology that facilitate rapid dispersal (Phillips et al., 2006; Brown et al.,

2014).

The data in the present study were gathered in the course of an investigation aimed at obtaining clutches of eggs from female toads from across the species’ Australian range, so that the offspring could be reared under standardised conditions to tease apart genetic vs. environmental contributions to the geographical divergence in toad dispersal behaviour

(Phillips et al., 2010b). We collected adult (> 90 mm snout-urostyle length [SUL]) toads from populations in long-colonised through to invasion-front areas, and maintained them under standardised conditions at an intermediate location (Northern Territory: 12°37’S, 131°18’E).

Because our aim was to obtain eggs, we attempted to breed toads as soon as possible (rather than at fixed intervals). Whenever we considered that an adult female might contain mature eggs (based on body shape), we injected her with a synthetic gonadotropin, leuprorelin acetate (Lucrin; Abbott Australasia) using 0.8 mL of Lucrin diluted 1:20 with saline to induce spawning. Simultaneously injected males were used to fertilise the clutches. This method offers a reliable way of assessing reproductive status: injected gravid females oviposit within

12 h, even in the absence of a male (Kouba & Vance, 2009). We injected some females more than once, in an attempt to produce multiple clutches, or if the first injection was unsuccessful.

The study was conducted twice, in 2007–2008 and in 2014. Within both studies, toads were collected from all localities in the late dry season (October–December), when females are most likely to contain eggs. We maintained toads from each population in captivity for similar periods, and injected them with Lucrin at similar intervals. The mean interval between capture and first injection differed between the two studies (overall mean 168 days; 355 days in 2007–2008, 119 days in 2014), as did our sampling locations. In 2007–2008, data were obtained from females from one long-colonised area (Cairns, 70 years post-invasion, N = 10 injections of nine females [where one female was injected twice]); two intermediate areas

(Normanton, 40 years post-invasion, N = 16 injections of nine females; Borroloola, 18 years post-invasion, N = 15 injections of nine females); and the invasion front (Timber Creek, <1 year post-invasion, N = 14 injections of four females). In 2014, we repeated the study with females from three long-colonised areas (>70 years post-invasion; eight injections of eight females from Innisfail; 19 injections of 15 females from Townsville; seven injections of six

16 females from Tully) and four invasion-front populations (<2 years post-invasion; 34 injections of 32 females from Wyndham; 41 injections of 32 females from El Questro; 24 injections of 13 females from Purnululu National Park; 10 injections of 10 females from

Oombulgurri). We recorded whether or not each toad laid eggs after injection and, in the

2007–2008 study, weighed females before and after laying to obtain estimates of relative clutch mass (RCM; the mass loss at oviposition divided by carcass mass: Shine, 1992). In the

2014 study, females were not weighed after laying.

Between sampling periods, the only major difference in methodology was that, in

2007–2008, animals were maintained outdoors in plastic containers (1 x 1 m) with a soil substrate and 10 toads per container. In 2014, toads were kept in larger enclosures (5 x 5 m), with 20–50 animals per enclosure. In both cases, animals were sex-segregated with water provided ad libitum. Artificial lights attracted insects to the enclosures, and these were supplemented with cockroaches, mealworms, and water-soaked dog biscuits, ensuring plentiful food supplies.

Our statistical analyses were performed using JMP, version 9.0 (SAS Institute, Cary,

NC) and R software (R Core Team, 2015) and included linear regression for comparisons of continuously distributed traits (e.g. population age vs. RCM or percentage of females laying), generalized linear mixed models for regression of binomial dependant variables (e.g. whether or not a female laid a clutch when injected, accounting for the random effect of sampling locality), and analyses of variance to compare mean values of continuous variables (e.g. the interval between successive injections) between range-core vs. invasion-vanguard populations.

Results

Reproductive Output Per Clutch

Linear regression detected no significant relationship between RCM and population age (r2 =

0.01, N = 24, P = 0.28; Cairns, 70 years post-invasion, N = 10; Normanton, 40 years post- invasion, N = 9; Borroloola, 18 years post-invasion, N = 5; Timber Creek, 0 years post- invasion, N = 1). Dividing the sample into toads from range-core vs. invasion-front populations yielded the same result (mean RCMs = 0.13 and 0.14, respectively; F1,23 = 0.01,

P = 0.91).

Reproductive Frequency

The proportion of adult females that laid eggs in response to their first injection was low for invasion-front animals but high in females from long-colonised areas. Overall, clutches were laid by approximately one-third of captive invasion-front females, whereas, under identical conditions, more than three-quarters of females from older populations laid eggs. Using a generalized linear mixed model (binomial error structure, logit link, subpopulation as a random effect), the effect of population age was highly significant (Z = 4.48, P < 0.0001;

(Figure 1). Other fixed effects in the model were: female SUL (Z = 2.98, P = 0.003), date of injection (Z = -0.28, P < 0.78), and duration of captivity (Z = 0.99, P = 0.33). The significant effect of maternal body size reflected a higher reproductive output in larger females.

Some females were injected multiple times, allowing us to calculate intervals between clutches. Mean intervals between injections were similar for older vs. invasion-front populations (2007–2008: 163 days vs. 110 days, F1,15 = 1.26, P = 0.28; 2014: 93 days vs. 87 days, F1,25 = 0.05, P = 0.82). No invasion-front females produced multiple clutches, whereas we obtained second clutches from six females from long-colonised areas (one in 2007–2008, five in 2014; comparing populations, contingency table χ2 = 10.89, d.f. = 1, P < 0.001; if we

18 include female body size [SUL], size effect χ2 = 0.11, d.f. = 1, P = 0.74, population effect χ2

= 12.58, d.f. = 1, P < 0.0005). The interval between clutches ranged from 33 to 151 days

(mean = 92.8, N = 5) in 2014; the females in 2007–2008 produced clutches 344 days apart.

Figure 1. Percentage of adult female cane toads (Rhinella marina) that laid eggs in captivity in response to synthetic gonadotrophin injection, as a function of population age (time since colonisation). Data are shown separately for two studies: one conducted in 2007–2008 and one in 2014. Captive toads were maintained at the same site in both studies.

Discussion

Reproductive frequency of captive toads showed a strong divergence associated with invasion history (Figure 1). Overall, larger female toads were more likely to lay eggs but, despite their smaller mean body size, female toads from long-established populations were far more likely to lay eggs in captivity, and sometimes produced multiple clutches. Although our sample size is small (and limited to the 2007–2008 data), reproductive output per clutch

(RCM) appears to be similar in the populations that we studied. That conservatism may reflect fecundity-independent costs of reproduction (e.g., risks incurred when travelling to spawning sites and breeding: Wells, 2010), so that natural selection favours delaying reproduction until a large clutch can be produced, irrespective of frequency (Bull & Shine,

1979).

The low reproductive frequency of captive invasion-front females is consistent with field observations, which document few clutches in water bodies at newly-colonised sites

(Crossland et al., 2008) and a scarcity of juveniles during early stages of establishment

(Brown et al., 2013). Thus, rapid evolution of dispersal at the cane toad invasion front is associated with a low rate of reproduction. Because our data are preliminary, and originally were gathered for a different purpose, we cannot unambiguously identify the processes responsible for that divergence. For example, deleterious mutations may accumulate at expanding range edges (Burton & Travis, 2008; Peischl et al., 2015), and the low reproductive output could be a result of non-adaptive accumulation of reproduction-reducing mutations. Alternatively, successive generations of interbreeding among the fastest- dispersing toads at the invasion front every generation (‘spatial sorting’: Shine et al., 2011) may have caused inbreeding effects. Another possibility is that environmental conditions experienced prior to capture (several months before egg-laying) influenced subsequent reproductive output, such that these geographical differences might result from

21 phenotypically plastic traits that are set early in life by a female’s experiences (perhaps, even during the larval stage). Lastly, the two sets of populations undoubtedly differed in other respects as well, which may have affected reproductive biology. For example, toads at the invasion front may tend to be younger (even though they are larger, on average), such that any ontogenetic changes in reproductive frequency are confounded with the geographical location of the population. However, we can confidently conclude that, after several months of captive maintenance under standard conditions, and with access to equal food supplies, invasion-front female toads were far less likely to contain mature eggs than were conspecifics from range-core populations.

Does the lower reproductive output of invasion-vanguard toads translate into lower individual fitness? Not necessarily because other factors differ also. The risk of predation may be higher at the invasion front (Phillips et al., 2008; Shine, 2010), although food is more abundant (Brown et al., 2013), parasites are rare (Phillips et al., 2010c), and low conspecific densities reduce rates of cannibalism for both larval and metamorph toads (Pizzatto & Shine,

2008; Crossland & Shine, 2012). Thus, even if invasion-front toads reproduce less often than those in long-established populations, they may achieve similar (or even greater) lifetime reproductive output because of a relative absence of density-dependent regulation.

Although the present study is preliminary, it suggests that invasion-front toads trade- off reproduction for dispersal. Because evolutionary processes acting in the vanguard of an invasion push dispersal rates upwards, invasion fronts facilitate the detection of trade-offs operating on dispersal (Ronce, 2007). Such trade-offs are critical for understanding not only evolution during range expansion (and consequent limits on rates of spread), but also the evolution of dispersal more generally (Clobert et al., 2001). Invasive species represent uniquely powerful systems in which to measure dispersal-related trade-offs.

Acknowledgements

We thank Martin Mayer, Tiago Dalcin, and Jayna DeVore for constructing enclosures;

Matthew Greenlees, Georgia Ward-Fear, Jodie Gruber, Kimberley Lemmen, and James

Smith for collecting toads; and the Department of Parks and Wildlife (Western Australia) for access to Purnululu National Park. Comments by two anonymous reviewers improved the manuscript. All procedures were approved by the University of Sydney Animal Ethics

Committee (L04/3-2006/3/4288 and 6705). The research was supported by the Australian

Research Council (DP0772418, DP0984771, DP1094646, FL120100074).

Literature Cited

Alford, R.A., Brown, G.P., Schwarzkopf, L., Phillips, B.L., Shine, R. 2009. Comparisons

through time and space suggest rapid evolution of dispersal behaviour in an invasive

species. Wildl. Res. 36: 23–28.

Belichon, S., Clobert, J., Massot, M. 1996. Are there differences in fitness components

between philopatric and dispersing individuals? Acta Oecol. 17: 503–517.

Bonte, D., Van Dyck, H., Bullock, J.M., Coulon, A., Delgado, M., Gibbs, M., Lehouck, V.,

Matthysen, E., Mustin, K., Saastamoinen, M., Schtickzelle, N., Stevens, V.M.,

Vandewoestijne, S., Baguette, M., Barton, K., Benton, T.G., Chaput-Bardy, A., Clobert,

J., Dytham, C., Hovestadt, T., Meier, C.M., Palmer, S.C.F., Turlure, C., Travis, J.M.

2012. Costs of dispersal. Biol. Rev. 87: 290–312.

Brown, G.P., Kelehear, C., Shine, R. 2013. The early toad gets the worm: cane toads at an

invasion front benefit from higher prey availability. J. Anim. Ecol. 82: 854–862.

Brown, G.P., Phillips, B.L., Shine, R. 2014. The straight and narrow path: the evolution of

straight-line dispersal at a cane-toad invasion front. Proc. R. Soc. B 281: 20141385.

Bull, J.J., Shine, R. 1979. Iteroparous animals that skip opportunities for reproduction. Am.

Nat. 114: 296–303.

Burton, O.J., Travis, J.M. 2008. The frequency of fitness peak shifts is increased at expanding

range margins due to mutation surfing. Genetics 179: 941–950.

Burton, O.J., Travis, J.M.J., Phillips, B.L. 2010. Trade-offs and the evolution of life-histories

during range expansion. Ecol. Lett. 13: 1210–1220.

Clobert, J., Danchin, E., Dhondt, A.A., Nichols, J.D. 2001. Dispersal. Oxford University

Press, Oxford, UK.

Cote, J., Clobert, J., Brodin, T., Fogarty, S., Sih, A. 2010. Personality-dependent dispersal:

characterisation, ontogeny and consequences for spatially structured populations. Phil.

Trans. R. Soc. B 365: 4065–4076.

Crossland, M.R., Brown, G.P., Anstis, M., Shilton, C., Shine, R. 2008. Mass mortality of

native anuran tadpoles in tropical Australia due to the invasive cane toad (Bufo marinus).

Biol. Conserv. 141: 2387–2394.

Crossland, M.R., Shine, R. 2012. Embryonic exposure to conspecific chemicals suppresses

cane toad growth and survival. Biol. Lett. 8: 226–229.

Karlsson, B., Johansson, A. 2008. Seasonal polyphenism and developmental trade-offs

between flight ability and egg laying in a pierid butterfly. Proc. R. Soc. B 275: 2131–

2136.

Kouba, A., Vance, C.K. 2009. Applied reproductive technologies and genetic resource

banking for amphibian conservation. Reprod. Fertil. Dev. 21: 719–737.

Kubisch, A., Fronhofer, E.A., Poethke, H.J., Hovestadt, T. 2013. Kin competition as a major

driving force for invasions. Am. Nat. 181: 700–706.

Lever, C. 2001. The Cane Toad: The History and Ecology of a Successful Colonist. Westbury

Academic and Scientific Publishing, Otley, West Yorkshire, UK.

Lindström, T., Brown, G.P., Sisson, S., Phillips, B.L., Shine, R. 2013. Rapid shifts in

dispersal behaviour on an expanding range edge. Proc. Natl. Acad. Sci. USA 110:

13452–13456.

Peischl, S., Kirkpatrick, M., Excoffier, L. 2015. Expansion load and the evolutionary

dynamics of a species range. Am. Nat. 185: E81–E93.

Perkins, T.A., Phillips, B.L., Baskett, M.L., Hastings, A. 2013. Evolution of dispersal and

life-history interact to drive accelerating spread of an invasive species. Ecol. Lett. 16:

1079–1087.

Phillips, B.L., Brown, G.P., Webb, J.K., Shine, R. 2006. Invasion and the evolution of speed

in toads. Nature 439: 803.

Phillips, B.L., Brown, G.P., Travis, J.M., Shine, R. 2008. Reid's paradox revisited: the

evolution of dispersal kernels during range expansion. Am. Nat. 172: S34–S48.

Phillips, B.L., Brown, G.P., Shine, R. 2010a. Life-history evolution in range-shifting

populations. Ecology 91: 1617–1627.

Phillips, B.L., Brown, G.P., Shine, R. 2010b. Evolutionarily accelerated invasions: the rate of

dispersal evolves upwards during the range advance of cane toads. J. Evol. Biol. 23:

2595–2601.

Phillips, B.L., Kelehear, C., Pizzatto, L., Brown, G.P., Barton, D., Shine, R. 2010c. Parasites

and pathogens lag behind their host during periods of host range-advance. Ecology 91:

872–881.

Pizzatto, L., Shine, R. 2008. The behavioral ecology of cannibalism in cane toads (Bufo

marinus). Behav. Ecol. Sociobiol. 63: 123–133.

R Core Team. 2015. R: A Language and Environment for Statistical Computing. R

Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/

Ronce, O. 2007. How does it feel to be like a rolling stone? Ten questions about dispersal

evolution. Annu. Rev. Ecol. Evol. Syst. 38: 231–253.

Ronce, O., Clobert, J. 2012. Dispersal syndromes. Pages 119–138 in Clobert, J., Baguette,

M.G., Benton, T.G., Bullock, J.M., eds. Dispersal Ecology and Evolution. Oxford

University Press, Oxford, UK.

Shine, R. 1992. Relative Clutch Mass and body shape in lizards and snakes: is reproductive

investment constrained or optimised? Evolution 46: 828–833.

Shine, R. 2010. The ecological impact of invasive cane toads (Bufo marinus) in Australia. Q.

Rev. Biol. 85: 253–291.

Shine, R., Brown, G.P., Phillips, B.L. 2011. An evolutionary process that assembles

phenotypes through space rather than through time. Proc. Natl. Acad. Sci. USA 108:

5708–5711.

Travis, J.M.J., Dytham, C. 2002. Dispersal evolution during invasions. Evol. Ecol. Res. 4:

1119–1129.

Urban, M.C., Phillips, B.L., Skelly, D.K., Shine, R. 2008. A toad more traveled: the

heterogeneous invasion dynamics of cane toads in Australia. Am. Nat. 171: E134–E148.

Wells, K.D. 2010. The Ecology and Behavior of Amphibians. University of Chicago Press,

London, UK.

Chapter 3

Athletic Anurans: The Impact of Morphology, Ecology, and Evolution on

Climbing Ability in Invasive Cane Toads

Manuscript published as:

Hudson, C.M., Brown, G.P., Shine, R. 2016. Athletic anurans: the impact of morphology, ecology and evolution on climbing ability in invasive cane toads. Biol. J. Linn. Soc. 119:

992–999.

Abstract

Although primarily terrestrial, cane toads (Rhinella marina) sometimes climb near-vertical surfaces (tree-trunks, cliffs, fences) during foraging or dispersal activities. We scored climbing ability (in laboratory trials) of 288 cane toads from four regions in Australia, plus two sites on the island of Hawai’i. We found strong divergence in climbing ability associated not only with sex and relative limb length, but also population of origin. Within each population, longer-limbed individuals (and hence, males rather than females) were better climbers, although the geographical divergence in climbing ability remained significant even when sex and limb length were included in multivariate regression models. The geographical difference in climbing ability (but not morphology) disappeared when the progeny were raised in captivity under identical conditions, without climbing opportunities. Although influenced by morphology, climbing ability in wild-caught cane toads appears to be driven primarily by local environmental conditions that facilitate and/or reward arboreal activity.

Key-words: Bufo marinus, climbing, invasive species, locomotion, morphology.

Introduction

Many animals use a variety of locomotor modes, even if they rely primarily on one or two.

For example, many birds walk and hop, as well as fly, and many lizards resort occasionally to bipedal rather than quadrupedal locomotion as an escape mechanism (Snyder, 1962). Even if a given locomotor mode is used only rarely, it may still be under significant selection: for example, it may enhance an individual’s ability to evade a predator, access a rare but highly profitable food resource or use an otherwise inaccessible shelter-site (Dickinson et al., 2000).

Hence, we might expect evolutionary pressures to generate geographical variation in an organism’s facility at using that infrequent locomotor mode, reflecting variation among populations in the nature and importance of situations in which that mode is useful.

Anurans exhibit substantial morphological diversity within a conservative body plan, and such diversity is tightly linked to the ecology and locomotor mode of a given species

(Gomes et al., 2009; Jorgensen & Reilly, 2013). Arboreal taxa (as in most hylids: Zug, 1978;

Marsh, 1994) are typically long-limbed and slender-waisted, whereas terrestrial anurans (as in most bufonids: Zug, 1978; Marsh, 1994) combine a heavyset body form with relatively short limbs (Gomes et al., 2009; Jorgensen & Reilly, 2013). Most anurans are capable of saltatory (‘leaping’) locomotion, although there is enormous interspecific variation in that ability (Gans & Parsons, 1966; Gomes et al., 2009; Vidal-García et al., 2014). On the ground, anurans can use a range of gaits, from leaping through to walking, crawling, hopping

(Anderson et al., 1991), and even bounding (Reilly et al., 2015). Arboreal species can readily switch to terrestrial locomotion, although the reverse situation is far more problematic. A terrestrial frog with short limbs and a thickset body is ill-suited to climbing and, to our knowledge, observations of ground-frogs climbing trees or cliffs are rare.

Given that background, we were surprised to observe invasive cane toads (Rhinella marina) (Figure 1A) climbing trees in Hawai’i (Figure 1B) and ascending steep cliffs in

29 tropical Australia (Figure 1C). In the former case, insect prey are more abundant in trees than on the ground (C.M. Hudson, pers. obs.); in the latter case, the toad invasion front is moving through a rocky and often hilly landscape. Given the plausible effect of climbing ability on an individual’s fitness in both of these circumstances, we conducted a study that aimed to quantify climbing ability in cane toads from a range of populations and explore the phenotypic determinants of climbing ability in this species. Cane toads at the invasion front travel much further and faster than do toads from range-core populations (Phillips et al., 2006,

2007; Alford et al., 2009; Llewelyn et al., 2010; Lindström et al., 2013). We predicted that invasion-front toads should be more adept at climbing because they likely more often encounter physical obstacles that can only be surmounted by climbing, especially in novel landscapes where alternative easier paths are unknown to the animal. This would be true even if the habitats occupied by invasion-front and range-core toads were similarly rugose. Within

Australia, we tested toads from two invasion-front populations: one range-core population and one of intermediate age. These four sites were compared with two long-colonised populations on the island of Hawai’i, where toads were introduced 3 years prior to the

Australian introduction. Hawai’ian toads are relatively sedentary (i.e. resembling range-core

Australian toads in this respect) but often forage on arboreal rather than terrestrial prey

(Ward-Fear et al., 2016). Thus, we predicted that Hawai’ian toads, similar to invasion-front

Australian toads, would be adept at climbing. A concurrent experiment in which progeny were being raised from a range of toad populations under common-garden conditions (with no opportunity to climb) offered the chance to investigate the degree to which geographical differences in climbing ability persist in such offspring (and, hence, the degree to which climbing ability is fashioned by an individual toad’s experiences rather than its inherited morphology).

Figure 1. Cane toads (Rhinella marina) climbing in laboratory and natural settings. A, male toad at the top of a vertical climbing cylinder during a trial to assess climbing ability. B, female toad climbing a tree trunk to catch falling insects at Wailoa River State Park, Hawai’i.

C, female toad among rocks at the Emma Gorge waterfall in Western Australia.

Materials and Methods

Specimen Capture and Collection Sites

Cane toads (R. marina) were brought from the Caribbean to Hawai’i in 1932 and from

Hawai’i to northeastern Australia in 1935 (Mungomery, 1935; Lever, 2001). Thus, all the populations that we studied sprang from a single colonising event (of 150 individuals) in

1932 (Lever, 2001). In 2014 and 2015, we collected 288 cane toads (mean ± SE snout-vent length [SVL] = 109.2 ± 0.6 mm) from locations across Australia and Hawai’i (Figs 2, 3).

Prior to a trial, toads were housed in 70-L plastic containers with a constant water supply to ensure proper hydration, and all trials were conducted within 96 h of capture to reduce the influence of stress from captivity. Individuals held in captivity for more than 48 h were provided with insects as food. We grouped collection sites into six major regions: Queensland

(QLD; 22 males, 26 females), New South Wales (NSW; 24 males, 24 females), Northern

Territory (NT; 22 males, 26 females), Western Australia (WA; 24 males, 24 females),

Hawai’i Wet (windward; 26 males, 22 females), and Hawai’i Dry (leeward; 28 males, 20 females).

Climbing Trials

Each toad was placed at the bottom of an open-topped mesh cylinder, and filmed for 2 h in an enclosed room. These chambers (anchored to the ground to prevent tipping) consisted of 2.5- cm plastic trellis mesh and measured 50 cm in height and 15 cm in diameter. We tested the climbing abilities of toads using these chambers to provide a standardised challenge that could be overcome only by climbing (rather than jumping upwards). In the wild, cane toads prefer anthropogenically disturbed environments (González-Bernal et al., 2016) and thus frequently encounter fences and other man-made obstacles. Trials were conducted between

18:00 h and 03:00 h, with 12 toads being tested simultaneously in adjacent cylinders (Figure

4). At the beginning and end of each trial, we measured ambient temperature using a handheld infrared thermometer, and calculated the mean ambient temperature during each trial. We scored toad behaviour from videos, as: (1) wait time before climbing (from the beginning of the trial); (2) time for a toad to reach the top of the chamber (starting as soon as both feet were off the ground); and (3) a binary measure (Y/N) of escape success. Toads were considered to have escaped the climbing cylinder if they reached the top (Figure 1A). Scores for ‘wait time’ and ‘climb time’ were only collected from toads that successfully completed the trial, and were log-transformed to normalise the data.

Common-Garden Offspring

We collected approximately 50 adult toads per site (25 males and 25 females) from three sites in northeastern QLD (Townsville, Innisfail, and Tully) and four sites in northern WA (El

Questro, Purnululu, Wyndham, and Oombulgurri) (for GPS locations and sex ratios, see

Figure 3; see also Supporting Information, Table S1) with the intention of producing five clutches per population. We induced spawning by injection of leuprorelin acetate (Lucrin;

Abott Australasia) using 1 mL of Lucrin diluted 1:20 with saline, and raised the resulting progeny under standard conditions in captivity (for details, see Phillips et al., 2010). All pairings were between males and females collected from within 50 km of each other. The resulting number of clutches was 31 (16 QLD, 15 WA). After metamorphs attained body lengths >20 mm, we toe-clipped them for identification and kept them in outdoor enclosures in groups of 30 (with mixed parental origins). We tested climbing ability (as above) of 72 individuals (37 QLD, 19 females, 18 males; 35 WA, 17 females, 18 males) when the toads were 18–20 months of age (mean SVL ± SE = 96.3 ± 1.2 mm).

Measurements and Statistical Analysis

Prior to climbing trials, we measured the body mass and SVL, as well as the length of the hand, radioulna, humerus, femur, tibiofibula, and foot of each toad. Toads were measured with Vernier callipers (± 0.1 mm) and weighed with spring scales (± 0.5 g). To compare relative limb lengths, we took the sums of individual measurements (e.g. hand, radioulna, and humerus) to obtain measurements of total arm length and total leg length. We obtained a measure of body condition for each toad by taking the residual scores from a linear regression of ln mass versus ln SVL. We then conducted a principal component (PC) analysis of SVL, total arm length, and total leg length measurements (Table 1). All three morphological variables were correlated positively with each other, and toads with relatively long arms also tended to have relatively long legs. Thus, we used PC axes to provide uncorrelated measures of size and relative limb length.

The first PC axis represented absolute body size (bodyPC), with larger values representing larger individuals. The second PC axis provides an index of limb length relative to body size (limbPC); as noted above, leg lengths and arm lengths were highly correlated.

To facilitate interpretation, we transposed the sign of limbPC values, such that positive values were indicative of a toad with longer limbs. We used linear and logistic regressions between morphological and performance variables to explore differences in climbing ability. We used two-factor analysis of variance (ANOVA) (with sex and population of parental origin as factors) to explore variation in morphology between common-garden toads from QLD versus

WA. Raw data for morphology and performance of all toads used in this study can be found in Table S2.

Figure 2. Representative habitat types where toads were collected. A, tropical rainforest in northeastern Australia (Mission Beach, Queensland). B, arid landscape in Purnululu National

Park, Western Australia. C, lava fields on the dry (leeward) side of the island of Hawai’i, near

Kona. D, rainforest on the wet (windward) side of the island of Hawai’i, north of Hilo.

Figure 3. Sampling locations within Australia (A) and Hawai’i (B). WA, Western Australia;

NT, Northern Territory; QLD, Queensland; NSW, New South Wales; HI DRY, Hawai’i Dry;

HI WET, Hawai’i Wet.

Figure 4. Vertical mesh cylinders used in trials to measure climbing ability in cane toads.

Results

Sexual Dimorphism in Wild-Caught Toads

Overall, male cane toads had longer limbs relative to SVL than did conspecific females (one- way ANOVA, N = 288, Nmale = 146, Nfemale = 142; F1,287 = 117.36, P < 0.0001). The same pattern was evident within every population, although the magnitude of dimorphism was variable (Figure 5).

Climbing Ability of Wild-Caught Toads

Logistic regression detected significant relationships between region, limbPC, and escape probability (Table 2), such that toads with higher limbPC scores were more likely to reach the top of the cylinder during climbing trials, and animals from some regions were more proficient at climbing than those from other regions (Figure 5). For example, >60% of

Hawai’ian male toads escaped during climbing trials, compared to <30% of females from

NSW, NT or the wet side of Hawai’i. Climbing time and wait time were significantly influenced by region of origin but not morphology (Table 2). Males had relatively longer limbs than females (see above) and, when we removed relative limb length from the model, sex became a strong predictor of climbing success (χ2 = 17.58, d.f. = 1,287, P < 0.0001).

Morphology and Climbing Ability of Common-Garden Progeny

Even when raised under identical conditions in captivity, toads varied substantially in relative limb length (the morphological trait that most strongly influenced climbing ability in wild- caught conspecifics). Two-factor ANOVA with sex and parental population origin as factors showed significant sex differences and a population * sex interaction for relative limb length

(PC2; sex, F1,68 = 21.6, P < 0.0001; population * sex, F1,68 = 5.83, P = 0.0185). However, our analyses on climbing ability (conducted as for the wild-caught sample but including only

38 common-garden individuals from QLD and WA populations) revealed no such sex or

2 2 geographical differences (sex, χ = 1.84, d.f. = 1,68, P = 0.17; population, χ = 0.84, d.f. =

1,68, P = 0.36). The analyses also revealed no significant sex or geographical differences in climb time (sex, F1,37 = 0.81, P = 0.11; population, F1,37 = 0.12, P = 0.53) or start time (sex,

F1,37 = 0.87, P = 0.36; population, F1,37 = 2.05, P = 0.16). Nonetheless, the influence of relative limb length on climbing ability remained. A univariate test indicated a strong effect of relative limb length on probability of escape (χ2 = 9.56, d.f. = 1,72, P = 0.002).

Table 1. Factor loadings, eigenvalues, and percentage variance explained by each principal component (PC) from principal component analysis of body size (snout-vent length [SVL]) and limb length in cane toads from several geographical locations.

Component Eigenvalues % Variance

PC 1 2.8475 94.918 PC 2 0.1214 4.046

Eigenvectors PC 1 PC 2

SVL 0.56836 0.80915 Total arm length 0.57903 -0.52215 Total leg length 0.58454 -0.26952

Table 2. The influence of body size [body principal component (PC)], body condition, region of origin, sex, temperature, and relative limb length (limbPC) on the escape probability of a cane toad (N = 288), ln escape time in seconds (N = 132), and ln start time in seconds (N =

132) in trials designed to measure climbing ability. Significant values (P < 0.05) are indicated in bold.

Trait Variable DF Test statistic P-value Escape probability bodyPC 1 χ2 = 0.009 0.92 (Y/N) Sex 1 χ2 = 2.78 0.10 limbPC 1 χ2 = 13.58 0.0002 Condition 1 χ2 = 2.05 0.15 Temperature 1 χ2 = 1.32 0.25 Region 5 χ2 = 20.47 0.0004 ln climb time bodyPC 1 F = 3.74 0.06 (seconds) Sex 1 F = 1.84 0.18 limbPC 1 F = 1.40 0.24 Condition 1 F = 0.93 0.34 Temperature 1 F = 1.01 0.32 Region 5 F = 3.64 0.0078 ln start time bodyPC 1 F = 0.06 0.80 (seconds) Sex 1 F = 1.84 0.09 limbPC 1 F = 1.39 0.12 Condition 1 F = 0.85 0.36 Temperature 1 F = 2.35 0.13 Region 5 F = 2.89 0.025

Figure 5. An adult cane toad’s likelihood of escape as a function of its relative limb length

(based on principal components analysis). The plot shows mean values for each sex from each geographic location (HI = Hawai’i, NSW = New South Wales, NT = Northern Territory,

QLD = Queensland, and WA = Western Australia).

Discussion

Belying their heavyset body and short stout limbs, cane toads (at least from some populations) are adept climbers. Plausibly, an individual capable of escaping from a pit (perhaps a space between grass clumps, a steep-sided pond or a hole in the ground) or climbing vegetation to access an arboreal food source might thereby enhance its viability (Figures 1B, 1C). More extensive field observations are needed to test that inference, although our discovery of strong geographical divergence in climbing ability supports the idea that this trait is phenotypically labile, and can be useful for a toad that is traversing complex environments.

Our prediction that invasion-front cane toads would possess superior climbing ability was not supported by the data; instead, our results suggest that local environmental conditions may determine the benefits of climbing and hence drive geographical variation in this trait.

The critical features include not only the overall physical attributes of the habitat, but also the distribution of prey resources. For example, long-colonised populations in QLD and Hawai’i inhabit similar habitats (parks, golf courses, rainforests) but behave differently as a result of local conditions of prey availability, ground cover, and soil moisture (Ward-Fear et al., 2016).

Much of the Hawai’ian landscape (especially on the western side of the island) is comprised of sharp, angular lava rock (Figure 2C), perhaps explaining the high climbing ability of

Hawai’ian dry-side males (>80% success in escape trials). On the other hand, invasion history also may play a role. Female toads from WA were the most successful climbers for their sex (>50% success in escape trials), which is consistent with our original prediction.

A toad’s ability to climb a vertical mesh cylinder was significantly enhanced by relatively longer limbs. Reflecting that functional link, male cane toads were better climbers than conspecific females (even from the same population: Figure 5). A tendency for longer limbs in males than females is widespread in anurans and generally has been attributed to sexual selection (Lee, 2001; Wells, 2010). For example, longer arms may enable a male to

43 more firmly grasp a female during amplexus and withstand efforts at displacement by rival males (Yekta & Blackburn, 1992; Peters & Aulner, 2000; Bowcock et al., 2013). In support of that hypothesis, Lee (1986) found that male cane toads in amplexus had longer forearms relative to SVL than did unsuccessful males. The occurrence of male-larger forelimbs in many anurans suggests that the greater climbing ability of males than females within R. marina may be a by-product of sexually-selected pressures, rather than having any adaptive significance in its own right. In keeping with such an interpretation, the strong sex divergence in climbing ability in our dataset disappears if we include relative limb length as a covariate in our analyses. That is, male toads appear to be better climbers than females simply because they have longer limbs, rather than for any other reason.

Even if climbing ability has not been a specific target of selection, sex-based divergences in limb structures may nonetheless have biologically significant implications for an individual’s mobility. More generally, sexually-selected disparities in traits such as mean adult body size doubtless influence mobility: for example, in some species of Anolis lizards, small adult females can access thin arboreal branches that are inaccessible to larger adult males, thereby gaining access to favoured food resources and providing an escape from male harassment (Jenssen et al., 1998; Rodríguez-Robles et al., 2005). More generally, sex differences in forelimb structure (apparently adaptive to male-male combat for grasping a female during mating) are very widespread, particularly in invertebrates (e.g. in male

Heteropterans, the front legs [tarsus] are modified to hold females: Popham et al., 1984; in

Thysanopterans, forelegs are enlarged for combat: Crespi, 1986; in Decapoda Natantia, males possess enlarged snapping claws: Schein, 1977; Knowlton, 1980; Mashiko, 1981; in many arachnid species, males have chitinous structures on chelicerae of the first leg to hold females: Hubert, 1979). Such morphological divergences, even if they have evolved primarily

44 in the context of sexual selection, may nonetheless influence relative locomotor abilities of the two sexes.

Regardless of relative limb lengths, some populations of cane toads were far better climbers than others (Figure 5). For example, 63% of toads from the leeward Hawai’i population succeeded in escaping from the mesh cylinder versus 29% of toads from the NT.

The trials on field-collected animals cannot evaluate the degree to which those differences were driven by adaptive (genetically-based) divergence versus plasticity. Plausibly, situations in which climbing allows a toad to access prey or traverse rocky landscapes may ‘train’ individuals to climb, thus generating location-specific differences in the ability to perform this task. By raising the offspring of toads from different populations under common conditions with no opportunity to climb, we found that geographical divergence in climbing ability is partly driven by a combination of heritable divergence in a morphological trait that enhances climbing (longer limbs). However, geographical divergence in climbing ability was no longer significant if relative limb length was included in the analysis, suggesting that local environmental conditions (presumably, the degree to which toads climb in nature, as well as the benefits they obtain by doing so) are responsible for much of the residual variation in climbing ability among toads from different locations.

In summary, cane toads are better climbers than we might have predicted from their morphology. Much of that variation in climbing ability is related to the sex of a toad, and likely is a secondary consequence of sexually-selected divergence in body shape (relative limb length). Even at similar limb lengths, however, toads from some populations are far more capable climbers than others. Such divergence may well reflect both heritable and phenotypically plastic responses to local conditions (as has been reported, in the same species, for traits related to hydric balance: Tingley et al., 2012; and also traits related to thermal tolerance: McCann et al., 2014), as well as evolutionary consequences of the invasion process

(as has been reported for dispersal rate, endurance, and immune function: Phillips et al., 2007;

Llewelyn et al., 2010; Brown et al., 2015). A diverse suite of phenotypic traits has diverged dramatically among populations of cane toads that shared a common ancestor only 80 years ago, and our data add climbing ability to that list. The breadth of traits exhibiting such divergence, and the twin roles of heritable and environmentally-induced variation, highlight the value of invasive species as model systems for exploring the process of rapid evolutionary change.

Acknowledgements

We thank Matthew Greenlees, Georgia Ward-Fear, Samantha McCann, Greg Clarke, Jodie

Gruber, and Georgia Kosmala for assistance with toad collection; Melanie Elphick for

formatting and editing the manuscript submitted for publication; and William Mautz for

information and resources in Hawai’i. Comments from three anonymous reviewers improved

the quality of this manuscript, and we thank them for their contributions. The present study

was funded by the Australian Research Council with approval from the State of Hawai’i

Department of Land and Natural Resources, Division of Forestry and Wildlife, and the

Animal Care and Ethics Committee of the University of Sydney (6705).

Literature Cited

Alford, R.A., Brown, G.P., Schwarzkopf, L., Phillips, B.L., Shine, R. 2009. Comparisons

through time and space suggest rapid evolution of dispersal behaviour in an invasive

species. Wildl. Res. 36: 23–28.

Anderson, B.D., Feder, M.E., Full, R.J. 1991. Consequences of a gait change during

locomotion in toads (Bufo woodhousii fowleri). J. Exp. Biol. 158: 133–148.

Bowcock, H., Brown, G.P., Shine, R. 2013. Sexual selection in cane toads (Rhinella marina): a

male’s body size affects his success and his tactics. Curr. Zool. 59: 747–753.

Brown, G.P., Phillips, B.L., Dubey, S., Shine, R. 2015. Invader immunology: invasion history

alters immune system function in cane toads (Rhinella marina) in tropical Australia. Ecol.

Lett. 18: 57–65.

Crespi, B.J. 1986. Territoriality and fighting in a colonial thrips, Hoplothrips pedicularius, and

sexual dimorphism in Thysanoptera. Ecol. Entomol. 11: 119–130.

Dickinson, M.H., Farley, C.T., Full, R.J., Koehl, M.A.R., Kram, R., Lehman, S. 2000. How

animals move: an integrative view. Science 288: 100–106.

Gans, C., Parsons, T.S. 1966. On the origin of the jumping mechanism in frogs. Evolution 20:

92–99.

Gomes, F.R., Rezende, E.L., Grizante, M.B., Navas, C.A. 2009. The evolution of jumping

performance in anurans: morphological correlates and ecological implications. J. Evol.

Biol. 22: 1088–1097.

González-Bernal, E., Greenlees, M.J., Brown, G.P., Shine, R. 2016. Toads in the backyard:

why do invasive cane toads (Rhinella marina) prefer buildings to bushland? Popul. Ecol.

58: 293–302.

Hubert, M. 1979. Les araignées. Societe Nouvelle des Editions Boubee, Paris.

Jenssen, T.A., Hovde, K.A., Taney, K.G. 1998. Size-related habitat use by nonbreeding Anolis

carolinensis lizards. Copeia 1998: 774–779.

Jorgensen, M.E., Reilly, S.M. 2013. Phylogenetic patterns of skeletal morphometrics and

pelvic traits in relation to locomotor mode in frogs. J. Evol. Biol. 26: 929–943.

Knowlton, N. 1980. Sexual selection and dimorphism in two demes of a symbiotic, pair-

bonding snapping shrimp. Evolution 34: 161–173.

Lee, J.C. 1986. Is the large-male mating advantage in anurans an epiphenomenon? Oecologia

69: 207–212.

Lee, J.C. 2001. Evolution of secondary sexual dimorphism in the toad, Bufo marinus. Copeia

2001: 928–935.

Lever, C. 2001. The Cane Toad: The History and Ecology of a Successful Colonist. Westbury

Academic and Scientific Publishing, Otley, West Yorkshire, UK.

Lindström, T., Brown, G.P., Sisson, S.A., Phillips, B.L., Shine, R. 2013. Rapid shifts in

dispersal behavior on an expanding range edge. Proc. Natl. Acad. Sci. USA 110: 13452–

13456.

Llewelyn, J., Phillips, B.L., Alford, R.A., Schwarzkopf, L., Shine, R. 2010. Locomotor

performance in an invasive species: cane toads from the invasion front have greater

endurance, but not speed, compared to conspecifics from a long colonised area.

Oecologia 162: 343–348.

Marsh, R.L. 1994. Jumping ability of anurans. Pages 51–111 in Jones, J.H. ed., Comparative

Vertebrate Exercise Physiology Academic Press, San Diego, CA, USA.

Mashiko, K. 1981. Sexual dimorphism of the cheliped in the prawn Macrobrachium

nipoinense (de Haan) and its significance in reproductive behaviour. Zool. Mag. (Tokyo)

90: 333–337.

McCann, S., Greenlees, M.J., Newell, D., Shine, R. 2014. Rapid acclimation to cold allows the

cane toad to invade montane areas within its Australian range. Funct. Ecol. 28: 1166–

1174.

Mungomery, R.W. 1935. The giant American toad (Bufo marinus). Cane Growers Q. Bull.

1935: 21–27.

Peters, S.E., Aulner, D.A. 2000. Sexual dimorphism in forelimb muscles of the bullfrog, Rana

catesbeiana: a functional analysis of isometric contractile properties. J. Exp. Biol. 203:

3639–3654.

Phillips, B.L., Brown, G.P., Webb, J.K., Shine, R. 2006. Invasion and the evolution of speed in

toads. Nature 439: 803.

Phillips, B.L., Brown, G.P., Greenlees, M., Webb, J.K., Shine, R. 2007. Rapid expansion of the

cane toad (Bufo marinus) invasion front in tropical Australia. Aust. Ecol. 32: 169–176.

Phillips, B.L., Kelehear, C., Pizzatto, L., Brown, G.P., Barton, D., Shine, R. 2010. Parasites

and pathogens lag behind their host during periods of host range-advance. Ecology 91:

872–881.

Popham, E.J., Bryant, M.T., Savage, A.A. 1984. The role of front legs of British carixid bugs

in feeding and mating. J. Nat. Hist. 18: 445–464.

Reilly, S.M., Montuelle, S.J., Schmidt, A., Naylor, E., Jorgensen, M.E., Halsey, L.G., Essner,

R.L. 2015. Conquering the world in leaps and bounds: hopping locomotion in toads is

actually bounding. Funct. Ecol. 29: 1308–1316.

Rodríguez-Robles, J.A., Leal, M., Losos, J.B. 2005. Habitat selection by the Puerto Rican

yellow chinned anole, Anolis gundlachi. Can. J. Zool. 83: 983–988.

Schein, H. 1977. The role of snapping in Alpheus heterochaelis (Say, 1818), the big-clawed

snapping shrimp. Crustaceana 33: 182–188.

Snyder, R.C. 1962. Adaptations for bipedal locomotion of lizards. Am. Zool. 2: 191–203.

Tingley, R., Greenlees, M.J., Shine, R. 2012. Hydric balance and locomotor performance of an

anuran (Rhinella marina) invading the Australian arid zone. Oikos 121: 1959–1965.

Vidal-García, M., Byrne, P.G., Roberts, J.D., Keogh, J.S. 2014. The role of phylogeny and

ecology in shaping morphology in 21 genera and 127 species of Australo-Papuan

myobatrachid frogs. J. Evol. Biol. 27: 181–192.

Ward-Fear, G., Greenlees, M.J., Shine, R. 2016. Toads on lava: spatial ecology and habitat use

of invasive cane toads (Rhinella marina) in Hawai’i. PLoS One 11: e0151700.

Wells, K. 2010. The Ecology and Behavior of Amphibians. The University of Chicago Press,

Chicago, IL, USA.

Yekta, N., Blackburn, D. 1992. Sexual dimorphism in mass and protein content of the forelimb

muscles of the northern leopard frog Rana pipiens. Can. J. Zool. 70: 670–674.

Zug, G.R. 1978. Anuran locomotion – structure and function, 2: jumping performance of

semiaquatic, terrestrial, and arboreal frogs. Smithson. Contrib. Zool. 276: 1–31.

Supporting Information

Supporting Information 1. Locations where toads were captured, showing number of individuals per site and GPS coordinates of each location.

Region Population # Toads # Males # Females Latitude Longitude Queensland Charters Towers 12 6 6 -20.050480 146.253815 Innisfail 12 6 6 -17.524681 146.032329 Townsville 12 6 6 -19.257627 146.817871 Tully 12 4 8 -17.932869 145.923556 New South Brooms Head 19 9 10 -29.608228 153.335818 Wales Tabbimobile 17 10 7 -29.199608 153.270409 Yamba 12 5 7 -29.437827 153.360272 Northern Fogg Dam 24 10 14 -12.568032 131.309507 Territory Katherine 12 6 6 -14.464967 132.264256 Pine Creek 12 6 6 -13.82484 131.834913 Western Kununurra 12 6 6 -15.773546 128.739196 Australia Purnululu 11 6 5 -17.529752 128.400838 Oombulgurri 12 6 6 -15.180417 127.845039 Wyndham 13 6 7 -15.464803 128.100143 Hawai’i Wet Richardson Park 12 8 4 19.736008 -155.013274 (Windward) Orchidland 12 7 5 19.557034 -155.017455 Wailoa River 12 6 6 19.720698 -155.07788 Panaewa Zoo 12 5 7 19.653754 -155.073765 Hawai’i Dry King’s Shops 12 6 6 19.916323 -155.88209 (Leeward) Kona Life Care 12 6 6 19.571108 -155.957051 Mauna Lani G.C. 24 16 8 19.942773 -155.862629

Supporting Information 2. Morphology and climbing performance data for all wild-caught

and captive-raised (common garden) cane toads.

Available online at:

http://onlinelibrary.wiley.com/store/10.1111/bij.12827/asset/supinfo/bij12827-sup-0002-

Supinfo.xlsx?v=1&s=4debdb0914769407425110f0679ad1b703868755

Chapter 4

Constructing an Invasion Machine: The Rapid Evolution of a Dispersal-

Enhancing Phenotype During the Cane Toad Invasion of Australia

Manuscript published as:

Hudson, C.M., McCurry, M.R., Lundgren, P., McHenry, C.R., Shine, R. 2016. Constructing an invasion machine: the rapid evolution of a dispersal-enhancing phenotype during the cane toad invasion of Australia. PLoS One 11: e0156950.

Abstract

Biological invasions can induce rapid evolutionary change. As cane toads (Rhinella marina) have spread across tropical Australia over an 80-year period, their rate of invasion has increased from around 15 to 60 km per annum. Toads at the invasion front disperse much faster and further than conspecifics from range-core areas, and their offspring inherit that rapid dispersal rate. We investigated morphological changes that have accompanied this dramatic acceleration, by conducting three-dimensional morphometric analyses of toads from both range-core and invasion-front populations. Morphology of heads, limbs, pectoral girdles and pelvic girdles differed significantly between toads from the two areas, ranging from 0.5% to 16.5% difference in mean bone dimensions between populations, with invasion-front toads exhibiting wider forelimbs, narrower hindlimbs and more compact skulls. Those changes plausibly reflect an increased reliance on bounding (multiple short hops in quick succession) rather than separate large leaps. Within an 80-year period, invasive cane toads have converted the basic anuran body plan – which evolved for occasional large leaps to evade predators – into a morphotype better-suited to sustained long-distance travel.

Key-words: Adaptation, Bufo marinus, dispersal phenotype, geometric morphometrics, locomotion, morphology, rapid evolution, cane toad.

Introduction

Biological invasions impose profound new evolutionary pressures both upon the invader, and upon the recipient ecosystem (Pimintel et al., 2000). In response to those pressures, organisms can exhibit phenotypic evolution at rates far higher than are usually observed in equilibrial systems (Thompson, 1998; Phillips & Shine, 2006). For example, individuals at an expanding range edge often exhibit distinctive traits of behaviour, physiology and morphology that enhance their rates of dispersal (Travis & Dytham, 2002; Simmons &

Thomas, 2004; Ronce & Clobert, 2012). The accumulation of dispersal-enhancing traits has been recorded at invasion fronts of organisms as diverse as pine trees (lighter seeds that float further on the wind: Cwynar & MacDonald, 1987), damselflies (larger wing musculature:

Therry et al., 2014), birds (larger wings: Berthouly-Salazar et al., 2012; Bitton & Graham,

2015) and rodents (larger feet: Forsman et al., 2011; see Chuang & Peterson, 2015 for a review).

One of the most intensively studied invasions is that of the cane toad (Rhinella marina) through tropical Australia (Shine, 2010). Introduced to northeastern Queensland in 1935 in a futile attempt to control insect pests, toads have spread at an ever-increasing pace (from 1–15 km/yr in the decades post-release, to 55–60 km/yr at present: Urban et al., 2007, 2008).

Radio-tracking studies confirm that range-core toads are sedentary (mean nightly displacement <10 m) whereas invasion vanguard toads are highly mobile (>200 m per night:

Phillips et al., 2007; Alford et al., 2009; Lindström et al., 2013). Laboratory-bred offspring raised in common-garden conditions inherit the distinctive dispersal rate (Phillips et al.,

2010), dispersal behaviour (path straightness: Brown et al., 2014), and immunological functioning (Brown et al., 2015) of their parents.

Has toad morphology also evolved in ways that facilitate rapid, sustained dispersal?

The anuran body plan is highly conservative, and centered around a powerful propulsive

54 system that can allow a frog to leap distances several times its own body length (Gans &

Parsons, 1966). That spectacular ability has been lost in many anuran lineages, especially fossorial taxa, but they retain the basic anuran morphotype of a large head, a short inflexible spinal column, and a lever system (involving the pelvic girdle and hindlimbs) that allows saltatory locomotion (Essner et al., 2010). In cane toads, invasion-vanguard toads were reported to have longer hindlimbs relative to body length than did the toads a year behind the front; and those longer legs were associated with more rapid dispersal (from radio-tracking:

Phillips et al., 2006; Alford et al., 2009) but also, with vulnerability to spinal arthritis (Brown et al., 2007; Shilton et al., 2008). Given the functional integration of body components, and the potential influence of many phenotypic traits on locomotor speed and endurance, we speculated that other morphological features might well have evolved also during the course of the toads’ Australian invasion. Accordingly, we conducted Computerized X-ray

Tomography (CT) scanning of toads to examine whether skeletal morphology varies between the long-colonised (eastern) and recently invaded (western) extremes of the species’ current distribution in Australia.

Materials and Methods

Study Species and Collection Sites

Between October and December, 2013 we collected 30 toads from two recently invaded populations in Western Australia (WA); El Questro Home Valley Station (16°00S, 127°580E) and Kununurra (15°460S, 128°440E). We also obtained 30 toads from a long-colonised population in Townsville, Queensland (QLD; 19°150S, 146°490E). The western sites were colonised by toads in 2012 (El Questro) and 2010 (Kununurra; Goodgame, 2015) while the eastern site was invaded in 1940 (Phillips et al., 2007), soon after toads were imported to

Australia. Following capture, these animals were humanely euthanized and shipped to

Melbourne, Victoria for imaging. We collected adults of both sexes, as well as juveniles to capture a range of body sizes for each population.

Imaging and Post-processing

From the initial 60 toads collected, 55 (QLD n = 27,WA n = 28) were used for scanning and geometric morphometric analysis (QLD: 16 males, 11 females, range 93.1 to 119.4 mm snout-vent length [SVL];WA: 13 males, 7 females, 8 not sexed, range 72.7 to 125.1 mm

SVL). The toads were scanned at the Melbourne Brain Center using a Siemens 128 slice

Computerised X-ray Tomography (CT) system. The resulting image stacks were imported into Mimics V16 software for data segmentation. Within each scan, each anatomical feature was digitally isolated and exported in polygon file format.

Landmarks and Geometric Morphometric Analysis

Landmarks were recorded using Landmark (version 3.0.0.6) software (Wiley, 2006) as three- dimensional Cartesian co-ordinates on the surface meshes. Figures and descriptions of the landmark locations are detailed in Figures A–H and Tables A–H in Supporting Information.

To eliminate size differences between individuals and to correct the dataset for translation and rotation we conducted a generalized Procrustes analysis in Morphologika (version 2.5;

O’Higgins & Jones, 1998), followed by a principal components analysis (PCA) to examine variation in shape. The first six principal components (PC) of each element were compared between populations using MANOVAs. Following this, each PC was then compared between populations and sexes with a one-way ANOVA. Significant PCs were also tested against centroid size (lnCS) to remove the effect of ontogeny on bone morphology. PCs where shape variation was more strongly linked to centroid size than to geographic origin were excluded.

Sexual dimorphism was minor, accounting for less than 5% of variation in shape. We

56 excluded PCs that were sexually dimorphic from comparisons between the populations, to avoid sample sex ratios confounding comparisons among areas. For simplicity, we report results from one-factor ANOVAs with area of origin as the factor, combining data from both sexes. For each PC axis that differed significantly between toads from eastern versus western

Australia, we produced visualisations of mean shape variation for invasion front and range- core individuals using EVAN toolbox V2.1 (Figures 1 to 7). Throughout this manuscript, figures containing visualisations were created using a hypothetical long-colonised toad as the reference (derived from the population mean shape), and a hypothetical invasion-front toad as the target shape. Visualisations reflect the shift in morphology from eastern to western toads.

To produce simplified estimates of the magnitude of difference between significant

PC values we compared the three-dimensional Cartesian co-ordinates from Landmark

(version 3.0.0.6) for specific regions of interest (e.g. total humerus length, points 1 and 6) for each individual, after correcting for overall size. Using the Pythagorean Theorem: distance2 =

(x2-x1)2 + (y2-y1)2 + (z2-z1)2 we calculated linear distances between points, and estimated the percent change in size between population means.

Results

The two populations did not differ significantly in mean SVL (108.5 vs. 102.2 mm in WA and QLD respectively; F1,53 = 3.43, P = 0.07), but differed strongly in morphology.

MANOVAs on each osteological element except the suprascapula detected significant differences between populations (Table 1). For each bone, post-hoc one-way ANOVAs detected at least one PC axis differing significantly between toads from the two areas (Table

2; and see Table I in Supporting Information for detailed descriptions of the influence of each

PC axis on bone morphology). Compared to range-core conspecifics, invasion-front cane toads had dorso-ventrally deeper skulls (+4.5%) with a wider inter-orbital distance (+6.3%;

Figure 1), and pectoral girdles that were more curved (+2.5–7.2%), with wider articulation surfaces at the gleohumoral joint (+7.5–16.5%; Figure 2). The humerus and radioulna of invasion-front toads were larger at the elbow (+2.7–5%; Figures 3 and 4), and the humerus was straighter and longer (+9.2%; Figure 3), whereas the radioulna was wider and less sharply angled at the ulnar end of the wrist (Figure 4). Invasion-front toads had a narrower pelvis (-0.5%), with a smaller pelvic area (-4.5%; Figure 5), and smaller heads on both the femur (-2.8–5.4%; Figure 6) and tibiofibula (-2.5–9.4%; Figure 7), with a decrease in total femur length (-8.1%; Figure 6). The tibiofibula also was larger at the knee, but smaller at the ankle, in invasion-front individuals, creating a difference in total length (-12.5%; Figure 7).

In summary, as cane toads have invaded across tropical Australia they have evolved substantial changes in skeletal morphology (more robust forelimbs, less robust hindlimbs, changes to the pectoral and pelvic girdles, and a narrower skull).

Table 1. Geographic divergence in shapes of bones in cane toads from western (invasion- front) and eastern (range-core) populations in Australia. The table shows MANOVA results from the first six principal component axes for each bone. Bone elements that are significantly different in shape between toads from Western Australia versus Queensland are highlighted in boldface.

Bone F-value df P-value

Skull 1.36 6,45 <0.0001

Pectoral girdle 2.56 6,15 0.002

Suprascapula 0.98 6,16 0.058

Humerus 2.07 6,12 0.018

Radioulna 3.00 6,14 0.001

Pelvic girdle 0.86 6,28 0.005

Femur 0.63 6,29 0.02

Tibiofibula 0.74 6,28 0.011

Table 2. Principal Components (PCs) representing statistically significant (P < 0.05) morphological divergences in shape between invasion-front (Western Australia [WA]) and range-core (Queensland [QLD]) populations of cane toads. The larger mean value for each

PC is highlighted in boldface font. Ranking of axes (in terms of variance explained) is calculated separately for each bone.

Bone PC % Variance QLD mean WA mean P-value

Skull 1 29.3 -0.01205 0.01205 0.004

Pectoral girdle 2 15.2 -0.01590 0.01590 0.004

Pectoral girdle 3 12.8 -0.01067 0.01067 0.045

Humerus 1 20.9 -0.01159 0.01288 0.018

Radioulna 3 11.8 0.00988 -0.01086 0.005

Radioulna 4 8.0 -0.00624 0.00687 0.041

Pelvic girdle 3 12.9 0.00473 -0.00361 0.018

Pelvic girdle 6 5.8 0.00271 -0.00361 0.046

Femur 3 8.8 0.00631 -0.00789 0.002

Tibiofibula 2 11.5 -0.00407 0.00542 0.014

Tibiofibula 5 7.0 0.00318 -0.00424 0.013

Figure 1. Differences in morphology of the skull between populations of cane toads, based on analyses of 52 specimens (26 Queensland [QLD], 26 Western Australia [WA]). Dorsal and lateral views depict mean skull morphology of toads from long-colonised areas (left, blue) and those from invasion-front populations (right, red). The central images overlay the ones on either side to reveal points of divergence, in this case reflecting the transformation from a low

(-0.06) to high (0.06) PC1 score.

Figure 2. Differences in morphology of the pectoral girdle between cane toads from long- colonised (left, blue) and invasion-front populations (right, red) for 22 specimens (11

Queensland [QLD], 11 Western Australia [WA]). The central image overlays the ones on either side to reveal points of divergence, in this case reflecting a transformation from a low

(-0.09) to high (0.06) PC2 score, and a low (-0.06) to high (0.06) PC3 score.

Figure 3. Differences in morphology of the humerus between populations of cane toads based on scans of 19 specimens (10 Queensland [QLD], 9 Western Australia [WA]). The images show mean values for cane toads from long-colonised (left, blue) and invasion-front populations (right, red). The central image overlays the ones on either side to reveal points of divergence, in this case reflecting a transformation from a low (-0.04) to high (0.04) PC1 score.

Figure 4. Differences in morphology of the radioulna between cane toads from two populations based on scans of 21 specimens (11 Queensland [QLD], 10 Western Australia

[WA]). Mean values for cane toads from long-colonised populations are shown on the left (in blue) and means for invasion-front populations on the right (in red). The central image overlays the ones on either side to reveal points of divergence, in this case reflecting a transformation from a high (0.04) to low (-0.04) PC3 score, and a low (-0.04) to high (0.02)

PC4 score.

Figure 5. Differences in morphology of the pelvic girdle between cane toads from two regions, based on scans of 35 specimens (20 Queensland [QLD], 15 Western Australia [WA]).

Dorsal and lateral views depict changes to mean pelvis morphology between toads from long- colonised areas (left, blue) and those from invasion-front populations (right, red). The central image overlays the ones on either side to reveal points of divergence. These images depict the transformation from a high (0.04) to low (-0.04) PC3 score, and a high (0.03) to low (-0.02)

PC6 score.

Figure 6. Differences in morphology of the femur between populations of cane toads for 36 specimens (20 Queensland [QLD], 16 Western Australia [WA]). Mean values for long- colonised populations are shown on the left (in blue) and those from invasion-front populations on the right (in red). The central images overlay the ones on either side to reveal points of divergence. These images depict the transformation from a high (0.04) to low (-0.04)

PC3 score.

Figure 7. Differences in morphology of the tibiofibula between cane toads from long- colonised (left, blue) to invasion-front populations (right, red), based on 35 specimens (20

Queensland [QLD], 15 Western Australia [WA]). The central images overlay the ones on either side to reveal points of divergence. These images depict the transformation from a low

(-0.04) to high (0.04) PC2 score, and a high (0.02) to low (-0.02) PC5 score.

Discussion

The rapid evolution of a high-dispersal phenotype of cane toads in Australia has been achieved via a remarkable divergence in skeletal morphology between individual toads from invasion-front versus range-core populations. These substantial changes (e.g. a 9.2% increase in humerus length and 12.5% decrease in tibiofibula length) represent rapid phenotypic evolution, as they have occurred over an 80-year period, within the span of a human lifetime.

Even more remarkably, those changes have occurred within a body plan that is otherwise highly conservative, not just within the >500 species of the Family Bufonidae (Pramuk,

2006), but even within the >6,500 species of anurans worldwide (Gans & Parsons, 1966).

Major adaptive radiations into distinctive niches (arboreal, aquatic, or fossorial ecotypes) have been associated with changes in overall anuran shape (especially of limb proportions:

Vidal-Garciá et al., 2014), but distantly-related anurans with similar ecological niches exhibit extensive morphological similarities (Wells, 2010; Moen et al., 2013). Indeed, that conservatism has been a major obstacle to phylogenetic analyses based on morphology (Frost et al., 2006).

The Bufonidae rapidly achieved a near-global distribution after originating in South

America and colonising North America, Eurasia and Africa between 78 to 98 Ma (Pramuk et al., 2007). This range expansion was primarily accomplished by toad species phenotypically similar to the cane toad (Van Bocxlaer et al., 2010). During its expansion across Australia, the cane toad has further elaborated these dispersal-enhancing morphological modifications.

Bufonids are more capable of sustained locomotion than are most other anurans, due to cardiovascular systems that can supply oxygen to active tissues over long periods (Bennett et al., 1973; Zug, 1985; Walton & Anderson, 1988; Anderson et al., 1991). Cane toads at the invasion front have been reported to show greater endurance than do conspecifics from range-core areas (Llewelyn et al., 2010; but see Tracey et al., 2012), plausibly reflecting

68 selection on this trait at the invasion front. To transform the bufonid body plan into a long- distance disperser, the other major changes required are to the locomotor apparatus.

Unlike other saltatory anurans that rely on maximising jump distance to escape predators (Essner et al., 2010), toad locomotion involves a combination of crawling and hopping (Emerson, 1978). Although their maximal jump distances are lower than those of many similarly-sized anurans, toads have evolved to use their forearms to absorb the shock of landing (Akella & Gillis, 2011). That role of the forearms has been expanded to support a novel locomotor mode that involves a cyclical hopping gait (hereafter, “bounding”) for rapid, sustained locomotion (Griep et al., 2013; Reilly et al., 2015). By eliminating the pause between successive leaps, a bounding toad can utilise the stored energy from compression of the limbs upon landing, to power the subsequent bound (Reilly et al., 2015)

Our data show that cane toads in the invasion vanguard exhibit larger forearms

(especially, wider joints), and smaller hindlimbs, with corresponding alterations to the pectoral and pelvic girdles. These changes suggest that toads at the invasion-front rely more on their forearms during dispersal – consistent with the biomechanical demands of sustained, cyclical hops. The morphological changes that have occurred over the course of the toads’ invasion have produced wider forearms (better able to absorb shock on landing) and a reduction in hindlimb power (to facilitate shorter bounds, rather than huge leaps). Although we have no data on the stresses imposed by the formidable athletic achievements of invasion- front toads, the high incidence of spinal arthritis in such animals (Brown et al., 2007; Shilton et al., 2008) hints that the changes we have recorded may include adaptations to reduce such stress (as well as to increase the energy efficiency or velocity of locomotion). The apparent contradiction between our results and those of Phillips et al. (2006) – decrease versus increase in relative hindlimb length – are due to curvilinearities in this trait. Hindlimb length

69 has decreased overall during the toad’s Australian invasion (current study), but is higher at the invasion front than in less-recently-colonised areas (Hudson et al., 2016).

The changes in skull shape are more difficult to interpret. Although the skull is not usually considered as a component of the locomotor system, the degree of facial tilt in

Leporids (Mammalia, Lagomorpha) correlates with locomotor mode, perhaps because changes to cranial structure can increase the visual field of the organism (Kraatz et al., 2015).

The shift in cranial morphology between QLD and WA toads may reflect an advantage of visual awareness in completing multiple rapid hopping and landing cycles. The increase in cranial height (plus the lateral skull compression in invasion-front toads) also may reduce the risk of injury to the brain from repeated take-offs and landings.

In the 80 years following their introduction to Australia, cane toads have expanded their range to an area greater than 1.2 million km2 (Urban et al., 2007). This expansion has occurred at an increasing rate, with the invasion front advancing more rapidly each year post- colonisation (Urban et al., 2007, 2008). In the process of evolving a rapid-dispersal phenotype, Australian R. marina have undergone substantial changes in skeletal morphology.

Those changes may have arisen either through natural selection (because faster dispersal enables individuals to exploit resource-rich areas before competitors arrive: Brown et al.,

2013) and/or spatial sorting (wherein traits that accelerate dispersal accumulate at an expanding range edge, regardless of fitness consequences: Shine et al., 2011). In the course of their Australian invasion, cane toads are not only changing the rate at which they move, but the way that they move as well. The distinctive morphology of the invasion-front toads suggests that they have shifted from a sedentary lifestyle that requires occasional hops, to one where they migrate westward by rapid, repeated bounding. Although skeletal morphology is conservative across anurans, the intense pressures stimulated in a biological invasion can

70 rapidly sculpt an organism’s morphology, as well as its physiology and behaviour, in ways that enable it to move further and faster than its ancestors.

Acknowledgements

We thank Chris Walmsley for assistance in collecting animals, Alistair Evans for the use of his EVAN toolbox software, Greg Brown for statistical advice, and Michelle Quayle for help in CT scanning.

Literature Cited

Akella, T., Gillis, G.B. 2011. Hopping isn't always about the legs: forelimb muscle activity

patterns during toad locomotion. J. Exp. Zool. A 315: 1–11.

Alford, R.A., Brown, G.P., Schwarzkopf, L., Phillips, B.L., Shine, R. 2009. Comparisons

through time and space suggest rapid evolution of dispersal behaviour in an invasive

species. Wildl. Res. 36: 23–28.

Anderson, B.D., Feder, M.E., Full, R.J. 1991. Consequences of a gait change during

locomotion in toads (Bufo woodhousii fowleri). J. Exp. Biol. 158: 133–148.

Bennett, A.F., Licht, P. 1973. Relative contributions of anaerobic and aerobic energy

production during activity in Amphibia. J. Comp. Physiol. 87: 351–360.

Berthouly-Salazar, C., van Rensburg, B.J., Le Roux, J.J., van Vuuren, B.J., Hui, C. 2012.

Spatial sorting drives morphological variation in the invasive bird, Acridotheris tristis.

PLoS One 7: e38145.

Bitton, P.P., Graham, B.A. 2015. Change in wing morphology of the European starling

during and after colonization of North America. J. Zool. 295: 254–260.

Brown. G.P., Kelehear, C., Shine, R. 2013. The early toad gets the worm: cane toads at an

invasion front benefit from higher prey availability. J. Anim. Ecol. 82: 854–862.

Brown, G.P., Phillips, B.L., Shine, R. 2014. The straight and narrow path: the evolution of

straight-line dispersal at a cane-toad invasion front. Proc. R. Soc. B 281: 20141385.

Brown, G.P., Phillips, B.L., Dubey, S., Shine, R. 2015. Invader immunology: invasion history

alters immune system function in cane toads (Rhinella marina) in tropical Australia.

Ecol. Lett. 18: 57–65.

Brown, G.P., Shilton, C., Phillips, B.L., Shine, R. 2007. Invasion, stress, and spinal arthritis

in cane toads. Proc. Natl. Acad. Sci. USA 104:17698–17700.

Chuang, A., Peterson, C.R. 2015. Expanding population edges: theories, traits, and trade-offs.

Glob. Change Biol. 22: 494–512.

Cwynar, L.C., MacDonald, G.M. 1987. Geographical variation of lodgepole pine in relation

to population history. Am. Nat. 129: 463–469.

Emerson, S.B. 1978. Allometry and jumping in frogs: helping the twain to meet. Evolution 32:

551–564.

Essner Jr. R.L., Suffian, D.J., Bishop, P.J., Reilly, S.M. 2010. Landing in basal frogs:

evidence of saltational patterns in the evolution of anuran locomotion.

Naturwissenschaften 97: 935–939.

Forsman, A., Merilä, J., Ebenhard, T. 2011. Phenotypic evolution of dispersal-enhancing

traits in insular voles. Proc. R. Soc. B 278: 225–232.

Frost, D.R., Grant, T., Faivovich, J., Bain, R.H., Haas, A., Haddad, C.F.B., De Sa´, R.O.,

Channing, A., Wilkinson, M., Donnellan, S.C., Raxworthy, C.J., Campbell, J.A., Blotto,

B.L., Moler, P., Drewes, R.C., Nussbaum, R.A., Lynch, J.D., Green, D.M., Wheeler,

W.C. 2006. The amphibian tree of life. Bull. Am. Mus. Nat. Hist. 297: 1–370.

Gans, C., Parsons, T.S. 1966. On the origin of the jumping mechanism in frogs. Evolution 20:

92–99.

Goodgame, D. 2015. Google Map of the Toad Movement in Western Australia. Kimberley

Toadbusters. URL http://www.canetoads.com.au/googmap.htm

Griep, S., Schilling, N., Marshall, P., Amling, M., Hahme, L.M., Haas, A. 2013. Pectoral

girdle movements and the role of the glenohumeral joint during landing in the toad,

Rhinella marina (Linnaeus, 1758). Zoomorphology 132: 325–338.

Hudson, C.M., Brown, G.P., Shine, R. 2016. It’s lonely at the front: Contrasting evolutionary

trajectories in male and female invaders. R. Soc. Open Sci. 3: 160687.

Kraatz, B.P., Sherratt, E., Bumacod, N., Wedel, M.J. 2015. Ecological correlates to cranial

morphology in Leporids (Mammalia, Lagomorpha). PeerJ 3: e844.

Lindström, T., Brown, G.P., Sisson, S., Phillips, B.L., Shine, R. 2013. Rapid shifts in

dispersal behaviour on an expanding range edge. Proc. Natl. Acad. Sci. USA 110:

13452–13456.

Llewelyn, J., Phillips, B.L., Alford, R.A., Schwarzkopf, L., Shine, R. 2010. Locomotor

performance in an invasive species: cane toads from the invasion front have greater

endurance, but not speed, compared to conspecifics from a long colonised area.

Oecologia 162: 343–348.

Moen, D.S., Irschick, D.J., Wiens, J.J. 2013. Evolutionary conservatism and convergence

both lead to striking similarity in ecology, morphology and performance across

continents in frogs. Proc. R. Soc. B 280: 20132156.

O'Higgins, P., Jones, N. 1988. Facial growth in Cercocebus torquatus: an application of three

dimensional geometric morphometric techniques to the study of morphological

variation. J. Anat. 193: 251–272.

Pimentel, D., Lach, L., Zuniga, R., Morrison, D. 2000. Environmental and economic costs of

nonindigenous species in the United States. Bioscience 50: 53–65.

Phillips, B.L., Brown, G.P., Greenlees, M., Webb, J.K. Shine, R. 2007. Rapid expansion of

the cane toad (Bufo marinus) invasion front in tropical Australia. Austral Ecol. 32:

169–176.

Phillips, B.L., Brown, G.P., Shine, R. 2010. Evolutionarily accelerated invasions: the rate of

dispersal evolves upwards during the range advance of cane toads. J. Evol. Biol. 23:

2595–2601.

Phillips, B.L., Shine, R. 2006. An invasive species induces rapid adaptive change in a native

predator: cane toads and black snakes in Australia. Proc. R. Soc. B 273: 1545–1550.

Phillips, B.L., Brown, G.P., Webb, J.K., Shine, R. 2006. Invasion and the evolution of speed

in toads. Nature 439: 803.

Pramuk, J.B. 2006. Phylogeny of South American Bufo (Anura: Bufonidae) inferred from

combined evidence. Zool. J. Linn. Soc. Lond. 146: 407–452.

Pramuk, J.B., Robertson, T., Sites, J.W., Noonan, B.P. 2007. Around the world in 10 million

years: biogeography of the nearly cosmopolitan true toads (Anura: Bufonidae). Glob.

Ecol. Biogeogr. 17: 72–83.

Reilly, S.M., Montuelle, S.J., Schmidt, A., Naylor, E., Jorgensen, M.E., Halsey, L.G., Essner,

R.L. 2015. Conquering the world in leaps and bounds: hopping locomotion in toads is

actually bounding. Funct. Ecol. 29: 1308–1316.

Ronce, O., Clobert, J. 2012. Dispersal syndromes. Pages 119–138. in Clobert, J., Baguette,

M.G., Benton, T.G., Bullock, J.M., eds., Dispersal Ecology and Evolution. Oxford

University Press, Oxford, UK.

Shilton, C.M., Brown, G.P., Benedict, S., Shine, R. 2008. Spinal arthropathy associated with

Ochrobactrum anthropi in free-ranging cane toads (Chaunus [Bufo] marinus) in

Australia. Vet. Pathol. 45: 85–94.

Shine, R. 2010. The ecological impact of invasive cane toads (Bufo marinus) in Australia. Q.

Rev. Biol. 85: 253–291.

Shine, R., Brown, G.P., Phillips, B.L. 2011. An evolutionary process that assembles

phenotypes through space rather than through time. Proc. Natl. Acad. Sci. USA 108:

5708–5711.

Simmons, A.D., Thomas, C.D. 2004. Changes in dispersal during species' range expansions.

Am. Nat. 164: 378–395.

Thompson, J.N. 1998. Rapid evolution as an ecological process. Trends. Ecol. Evol. 13: 329–

332.

Therry, L., Nilsson-Oertman, V., Bonte, D., Stoks, R. 2014. Rapid evolution of larval life

history, adult immune function and flight muscles in a poleward-moving damselfly. J.

Evol. Biol. 27: 141–152.

Tracey, C.R., Christian, K.A., Baldwin, J., Phillips, B.L. 2012. Cane toads lack physiological

enhancements for dispersal at the invasive front in Northern Australia. Biol. Open 1:

37–42.

Travis, J.M.J., Dytham, C. 2002. Dispersal evolution during invasions. Evol. Ecol. Res. 4:

1119–1129.

Urban, M.C., Phillips, B.L., Skelly, D.K., Shine, R. 2007. The cane toad's (Chaunus [Bufo]

marinus) increasing ability to invade Australia is revealed by a dynamically updated

range model. Proc. R. Soc. B 274: 1413–1419.

Urban, M., Phillips, B.L., Skelly, D.K., Shine, R. 2008. A toad more traveled: the

heterogeneous invasion dynamics of cane toads in Australia. Am. Nat. 171: 134–148.

Van Bocxlaer, I., Loader, S.P., Roelants, K., Biju, S.D., Menegon, M., Bossuyt, F. 2010.

Gradual adaptation toward a range-expansion phenotype initiated the global radiation

of toads. Science 327: 679–682.

Vidal-Garciá, M., Byrne, P.G., Roberts, J.D., Keogh, J.S. 2014. The role of phylogeny and

ecology in shaping morphology in 21 genera and 127 species of Australo-Papuan

myobatrachid frogs. J. Evol. Biol. 27: 181–192.

Walton, M., Anderson, B.D. 1988. The aerobic cost of saltatory locomotion in the Fowler's

toad (Bufo woodhousei fowleri). J. Exp. Biol. 136: 273–288.

Wells, K. 2010. The Ecology and Behavior of Amphibians. The University of Chicago Press,

Chicago, IL.

Wiley, D. 2006. Landmark Editor 3.0. Institute for Data Analysis and Visualization,

University of California, Davis, CA.

Zug, G.R. 1985. Anuran locomotion: fatigue and jumping performance. Herpetologica 41:

188–194.

Supporting Information

S1 File. Landmarks and descriptions for bone elements examined in this study: Skull (Figure

A, Table A), pectoral girdle (Figure B, Table B), suprascapula (Figure C, Table C), humerus

(Figure D, Table D), radioulna (Figure E, Table E), pelvic girdle (Figure F, Table F), femur

(Figure G, Table G), and tibiofibula (Figure H, Table H). Also included are detailed descriptions of the effect of a large mean value on bone shape for significant Principal

Component (PC) axes (Table I).

S1 Figure A and Table A

Skull Landmarks 1 Occipital condyle

3 Posterior squamosal process 5 Posterior edge of quadratojugal bone

7 Anterior squamosal process (next to orbit) 9 Anterior edge of fossa between pterygoid and quadratojugal 11 Widest point of curve on fronto-parietal bone 13 Upper curve of nasal bone (dorsal) 15 Lower curve of nasal bone (dorsal) 17 Terminal edge of premaxillary bone 19 Top of dentary bone

S1 Figure B and Table B

Pectoral Girdle Landmarks 1 (Scapula) Anterior articulation point with suprascapula 2 Highest point on connecting cartilage between clavicle and scapula 3 Lowest point in groove of clavicle 4 Terminal end of clavical (sternoclavicular joint) 5 Mid-point on groove between clavicle and coracoid 6 Anterior process of coracoid (articulation point with sternum) 7 Posterior process of coracoid (articulation point with sternum) 8 Proximal ridge of glenohumeral joint 9 Medial ridge of glenohumeral joint 10 Distal ridge of glenohumeral joint 11 Dorsal aperture of glenohumeral fossa 12 Ventral aperture of glenohumeral fossa 13 Distal aperture of glenohumeral fossa 14 Proximal aperture of glenohumeral fossa 15 (Scapula) Posterior articulation point with suprascapula

S1 Figure C and Table C

Suprascapular Landmarks 1 Anterior distal edge, at articulation point with scapula 2 Anterior proximal edge of bone 3 Posterior proximal edge of bone 4 Mid-point of curvature between points #3 and #4 5 Posterior distal edge – opposite point #6 6 Posterior distal edge, at articulation point with scapula 7 Highest point on cleithrum

S1 Figure D and Table D

Humerus Landmarks 1 Terminal end of posterior extremity (cubital joint) 2 Lateral edge of posterior extremity (trochlear side) 3 Opposite point #2 (non-trochlear side) 4 Doral edge of posterior extremity

5 Opposite point #4 (ventral) 6 Terminal end of anterior extremity (glenohumeral joint)

7 Lateral edge of anterior extremity (non-trochlear side) 8 Opposite point #7 (trochlear side) 9 Dorsal edge of anterior extremity 10 Opposite point #9 (ventral) 11 Anterior edge of crista deltoidea 12 Mid-point of crista deltoidea 13 Posterior edge of crista deltoidea 14 Lateral mid-point of bone (non-trochlear side) 15 Opposite point #12 16 Opposite point #14 17 Lateral edge of crista medialis (where it meets head of humerus) 18 Opposite point # 17 19 Groove under head at cubital joint

S1 Figure E and Table E

Radioulna Landmarks 1 Centre of greater sigmoid cavity (cubital joint)

2 Capitulum radii 3 Lower ridge of olecranon 4 Upper ridge of olecranon 5 Terminal end, at groove between ulna and radius (radiocarpal joint) 6 Lateral (ulnar) terminal end 7 Lateral (radial) terminal end 8 Dorsal terminal end 9 Ventral terminal end 10 Dorsal midpoint of bone 11 Lateral (ulnar) midpoint 12 Opposite point #10 (ventral) 13 Opposite point #11 (radial)

S1 Figure F and Table F

Pelvic Girdle Landmarks 1 Terminal point of left iliac arm 2 Superior process of left ilium 3 Meeting point between iliac arms 4 Tuber ischii 5 Inferior process of the ilium 6 Left ischial acetabular margin 7 Left pubic acetabular margin 8 Right ischial acetabular margin 9 Right pubic acetabular margin 10 Superior process of right ilium 11 Terminal point of right iliac arm

S1 Figure G and Table G

Femur Landmarks 1 Terminal end of posterior extremity (patellofemoral joint)

2 Lateral end of posterior extremity (left side) 3 Opposite point #3 4 Dorsal end of posterior extremity (top of knee) 5 Opposite point #5 (ventral) 6 Terminal end of anterior extremity (acetabular joint) 7 Lateral end of anterior extremity (right side) 8 Opposite point #7 9 Dorsal end of anterior extremity 10 Opposite point #9 (ventral) 11 Anterior edge of crista femoralis 12 Mid-point of crista femoralis 13 Posterior edge of crista femoralis 14 Lateral mid-point of bone (right side) 15 Opposite point #12 (dorsal) 16 Opposite point #14

S1 Figure H and Table H

Tibiofibula Landmarks 1 Terminal edge of anterior extremity (patellofemoral joint) 2 Lateral end of anterior extremity (left side) 3 Opposite point #2 4 Terminal edge of posterior extremity (articulation point with tarsus) 5 Lateral edge of posterior extremity (same side as calcaneum) 6 Opposite point #5 (same side as astragalus) 7 Dorsal end of anterior extremity 8 Opposite point #7 (inside of knee) 9 Dorsal end of posterior extremity (inside of ankle) 10 Opposite point #9 11 Mid-point of bone (outside edge) 12 Lateral midpoint (right side) 13 Opposite point #11 (inside edge) 14 Opposite point #12

S1 Table I

Bone PC Description (of large PC value) Axis Skull 1 Increased cranial height, shortening of distance between posterior squamosal processes (points 3 and 4) and fronto-parietal curve above orbit (points 11 and 12), resulting in lateral skull compression. Pectoral 2 Increased aperture of glenohumeral fossa (points 11-14). girdle Decreased height of connecting cartilage between clavicle and scapula (point 2). Pectoral 3 Increased curvature of pectoral girdle, larger girdle glenohumeral joint (points 8, 9 and 10). Humerus 1 Increased bone straightness (angle between points 1 and 6). Larger head of bone at cubital joint (increased distance between points 4 and 5). Radioulna 3 Compression of cubital joint (decreased distance between 2, 3 and 4). Increased height of bone at radiocarpal joint (points 8 and 9). Radioulna 4 Increased distance between points 2 and 4 (at cubial joint). Ulnar end of radiocarpal joint (point 6) less angular, resulting in flattening at wrist. Pelvic girdle 3 Increased distance and angle between iliac arms (points 1,11). Inferior process of ilium upturned (decreased distance between points 3 and 5). Pelvic girdle 6 Broadening of acetabulum, increased height of superior iliac processes (points 2, 10), decreased height of tuber ischium. Femur 3 Heads of bone larger at patellofemoral (points 1-5) and acetabular joints (points 6-10) Tibiofibula 2 Heads of bone smaller at patellofemoral (points 1,2,3,7,8) and tarsal joint (points 4,5,6,9,10). Tibiofibula 5 Reduced size of patellofemoral head, expressed as increased distance of point 1 from midpoint

Chapter 5

It’s Lonely at the Front: Contrasting Evolutionary Trajectories in Male and

Female Invaders

Manuscript published as:

Hudson, C.M., Brown, G.P., Shine, R. 2016. It’s lonely at the front: contrasting evolutionary trajectories in male and female invaders. Roy. Soc. Open Sci. 3: 160687.

Abstract

Invasive species often exhibit rapid evolutionary changes, and can provide powerful insights into the selective forces shaping phenotypic traits that influence dispersal rates and/or sexual interactions. Invasions also may modify sexual dimorphism. We measured relative lengths of forelimbs and hindlimbs of more than 3000 field-caught adult cane toads (Rhinella marina) from 67 sites in Hawai’i and Australia (1–80 years post-colonisation), along with 489 captive-bred individuals from multiple Australian sites raised in a ‘common garden’ (to examine heritability and reduce environmental influences on morphology). As cane toads spread from east to west across Australia, the ancestral condition (long limbs, especially in males) was modified. Limb length relative to body size was first reduced (perhaps owing to natural selection on locomotor ability), but then increased again (perhaps owing to spatial sorting) in the invasion vanguard. In contrast, the sex disparity in relative limb length has progressively decreased during the toads’ Australian invasion. Offspring reared in a common environment exhibited similar geographical divergences in morphology as did wild-caught animals, suggesting a genetic basis to the changes. Limb dimensions showed significant heritability (2–17%), consistent with the possibility of an evolved response. Cane toad populations thus have undergone a major shift in sexual dimorphism in relative limb lengths during their brief (81 years) spread through tropical Australia.

Key-words: Bufo marinus, Rhinella marina, evolution, invasive species, morphology, sexual dimorphism.

Introduction

Invasive species offer unparalleled opportunities to explore the process of rapid evolutionary change (Moran & Alexander, 2014). As an alien species spreads through previously uncolonised territory, it is likely to encounter novel selective forces (both biotic and abiotic); and the process of continuous range expansion introduces an additional set of evolutionary processes (e.g., genetic drift, mutation surfing, spatial sorting: Shine et al., 2011). As a result, alien taxa often accumulate substantial phenotypic changes, at a timescale much quicker than usually envisaged for evolutionary change (Moran & Alexander, 2014; Chuang & Peterson,

2015; Rollins et al., 2015). One interesting subset of traits that might be expected to evolve during a biological invasion involves sexually dimorphic characteristics. Morphological disparities between conspecific males and females take many forms (Andersson, 1994), but evolutionary theory suggests that natural selection during an invasion might act most forcefully on traits that affect dispersal rate and/or reproductive characteristics (Chuang &

Peterson, 2015).

Dispersal rate is a key feature of invasion biology, and an extensive literature suggests that dispersal rate typically evolves upwards during an invasion (Travis & Dytham, 2002;

Simmons & Thomas, 2004; Phillips et al., 2007, 2008; Ronce & Clobert, 2012). As a result, individuals in the invasion vanguard tend to exhibit dispersal-enhancing features (such as seeds that drift further on the wind, larger feet, wings or flight muscles: Cwynar &

MacDonald, 1987; Forsman et al., 2011; Berthouly-Salazar et al., 2012; Therry et al., 2014;

Bitton & Graham, 2015; Chuang & Peterson, 2015) relative to conspecifics in the range-core.

Alleles that code for fast-dispersal morphological traits may accumulate in the invasion vanguard because of spatial sorting (successive generations of interbreeding between the fastest dispersers: Shine et al., 2011) or natural selection (reflecting fitness benefits to unusually fast-moving individuals: Brown et al., 2013). However, what happens to this

89 acceleration if males and females within a population differ in traits (body size, limb dimensions, wing size, frequency of winged morph, etc.) that influence dispersal rate? If one sex is intrinsically faster than the other, individuals of that sex may be under intense counter- selection against more rapid dispersal, because they would encounter no potential mates at the expanding range-edge (Miller et al., 2011). Thus, we might expect to see an evolutionary reduction in the magnitude of sexual dimorphism in traits that enhance dispersal rate.

Sexual selection plays a critical role in the evolution of sexually dimorphic traits, reflecting the way that specific morphological features affect individual reproductive success

(Andersson, 1994; Arnqvist & Rowe, 2005). As a result, geographical differences in mating systems within widespread species often are associated with variation in patterns of sexual dimorphism (Slip & Shine, 1988; Pearson et al., 2002). By definition, population density is low at an invasion front (density is zero immediately in advance of the front) and, hence, individuals in the invasion vanguard may face very different rates of encounter with mates and consexual rivals than would be the case in the range-core (Kokko & Rankin, 2006). That disparity would be exacerbated by any shifts in the operational sex ratio between the range- edge and range-core (Kokko & Rankin, 2006). In circumstances where mates are rare, individual fitness will depend on traits that promote survival and longevity rather than success in mate competition (Martin & Hosken, 2003; Carranza & Pérez-Barbería, 2007;

Bonduriansky et al., 2008; Travers et al., 2015). In contrast, high population density can increase intrasexual competition and result in the evolution of secondary sexual characters such as ornaments, weaponry or sexual size dimorphism (Shine, 1979; Andersson, 1994;

Arnqvist & Rowe, 2005; Hudson & Fu, 2013). This is especially true for the sex in which reproductive success is limited by access to mates (typically males: Kokko & Rankin, 2006), and stems from the disparity in genetic interests between the sexes (‘sexual conflict’:

Chapman et al., 2003; Arnqvist & Rowe, 2005). Thus, both the ‘dispersal rate’ and ‘sexual

90 selection’ hypotheses predict the evolution of reduced sexual dimorphism at an expanding range-edge.

To test this prediction, we gathered data on cane toads (Rhinella marina; formerly

Bufo marinus) from 67 populations across the species’ invaded range in Hawai’i and tropical

Australia. Importantly, dates of introduction or arrival at each of these sites are well documented, allowing us to explore shifts in sexual dimorphism as a function of time since colonisation (TSC). Densities of invasive cane toad populations vary dramatically with TSC

(Freeland, 1986). Comparisons between range-edge and range-core toad populations within tropical Australia have demonstrated substantial shifts in morphology (Phillips et al., 2006;

Hudson et al., 2016a), locomotion (Llewelyn et al., 2010; Hudson et al., 2016b), dispersal ability (Phillips et al., 2007; 2008; 2010a; Alford et al., 2009; Shine et al., 2011; Lindström et al., 2013; Brown et al., 2014), hydric and thermal tolerance (Tingley et al., 2012; McCann et al., 2014), immune function (Brown et al., 2015), life-history traits (Phillips et al., 2010b), reproductive frequency (Hudson et al., 2015), incidence of spinal arthritis (Brown et al., 2007) and larval plasticity (Ducatez et al., 2016). Thus, cane toads are a plausible candidate species for examining invasion-associated shifts in sexual dimorphism.

We chose to focus on length of forelimbs and hindlimbs relative to the toad’s snout– vent (body) length (SVL). As well as being straightforward to measure accurately, limb dimensions have been reported to influence mating success of male cane toads (Lee, 2001;

Lee & Corrales, 2002), and toads with longer hindlimbs travel further distances daily when radio-tracked (Phillips et al., 2006). Thus, relative limb length is a likely target of natural selection and spatial sorting as well as sexual selection. Additionally, our preliminary analyses revealed significant sexual dimorphism in this trait: male cane toads have longer limbs (relative to body length) than do conspecific females, a pattern that is also widespread in other amphibians (Wells, 2010).

Simply documenting geographical variation in a trait (or dimorphism in that trait) does not establish any evolutionary causation, however. Anuran amphibians (including cane toads) exhibit high levels of developmental plasticity. The conditions in a water body

(temperature, food supply, competitor density, predator cues, etc.) can significantly affect the morphological traits of toads that metamorphose from that site (Hagman et al., 2009;

Wijethunga et al., 2016). In at least some anurans, rearing conditions of tadpoles can influence relative limb lengths of metamorphs (Relyea, 2001; Van Buskirk & Saxer, 2001;

Relyea & Hoverman, 2003; Alho et al., 2011). To sustain an explanation couched in terms of evolution rather than developmental plasticity, we need to breed toads in captivity, and raise their offspring to maturity in standard conditions. Unless offspring resemble their parents in relative limb length, and geographical divergences in morphology among wild-caught animals are also seen in their offspring even after the latter are raised in standardised conditions, we cannot attribute divergences among populations to evolutionary forces rather than to plasticity. Quantifying heritability also allows us to determine if morphological traits are likely to be able to respond to selection. We thus also conducted a breeding study.

Materials and Methods

Capture of Specimens and Sites of Collection

From August 2013 to March 2016, we collected adult cane toads from sites along a transect in tropical Australia (total N = 2076; Queensland [QLD] N = 467, Northern Territory [NT] N

= 802, Western Australia [WA] N = 807). We also collected toads from three Hawai’ian islands (N = 1018; Hawai’i N = 529, O’ahu N = 333, Maui N = 156) between January and

July of 2015. Toads were captured by hand, and we used vernier callipers (±0.1mm) to measure the SVL and limb lengths of each toad. The hand, radioulna and humerus were measured for the forelimb (arm), whereas the femur, tibiofibula and foot were measured for

92 the hindlimb (leg). Values for each component of a limb were added together to obtain measures of total forelimb length and total hindlimb length. Toads were sexed by examining external morphological characteristics (e.g., males possess nuptial pads on the thumbs, rugose dorsal skin and yellow colouration) and vocalisations (e.g., males observed calling, or producing release calls upon handling). Most individuals greater than 90 mm in SVL are sexually mature (Alford et al., 1995).

Common Garden Offspring

We collected a subgroup of adult Australian cane toads (approx. 25 males and 25 females per population) from the two extremes of the invaded range to conduct a ‘common garden’ breeding experiment. These toads were collected from three long-established populations in northeastern Queensland (more than 70 years since colonisation; Townsville, Innisfail, Tully) and four recently colonised sites in northern Western Australia (less than 3 years since colonisation; El Questro, Purnululu, Wyndham, Oombulgurri). Using protocols outlined by

Phillips et al. (2010c), we induced breeding pairs to spawn by injection of leuprorelin acetate

(Lucrin; Abbott Australasia, Botany, NSW) using 1 ml of Lucrin diluted 1:20 with saline and raised the resulting progeny in captivity at our field station in the Northern Territory (12°37'S,

131°18' E). All dams and sires were bred only once in this study, thus F1 individuals from each clutch were full-siblings. Once metamorphic toads attained SVLs more than 20 mm, we toe-clipped them for identification and moved them into outdoor enclosures in groups of 30

(with mixed parental origins). As F1 toads grew, they were split into smaller groups (approx.

10 by adulthood) to reduce competition for food and space, and avoid cannibalism. From this common garden study, we obtained data on 489 captive-raised offspring (287 Queensland,

202 Western Australia) from 31 egg clutches (16 Queensland, 15 Western Australia). We measured the same traits on these offspring as we did on wild toads. Offspring were

93 measured at approximately 2, 8, 14 and 17 months of age to quantify changes in skeletal morphology with growth and maturity. By their fourth measurement, 184 individuals had reached maturity and could be sexed based on the same criteria as used for wild-caught toads

(see above).

Statistical Analyses

We used multiple regressions to assess the effects of sex and TSC (in categories of 0–10 years, 11–20 years, 41–50, 51–60, 71–80 years) on relative arm and leg lengths of wild toads.

The SVL was included as a covariate in the models to control for body size. For graphical purposes, we calculated % limb length by dividing arm and leg measures by SVL. We calculated the difference in mean values of % limb lengths (for arms and legs separately) of males and females as an index of sexual dimorphism. To estimate the repeatability of our measurements of toad limbs, on one occasion, we took triplicate measures from five individuals. We used the R package rptR (Schielzeth et al., 2016) to calculate repeatability of measures of each limb component.

We used linear-mixed models to determine whether the relative length of limbs (and their components) of male and female toads raised in a common environment were affected by their parents’ location of origin (long-established populations in QLD versus invasive populations in WA). State and sex and their interaction were included as fixed effects in the models, along with SVL to control for body size. Individual ID (nested within clutch and state) and clutch (nested within state) were included as random effects in the models. This analysis was conducted on the final measurements made on the 184 individuals that were mature at the end of the study (76 F, 108 M).

To compare patterns of relative limb length of toads raised under common garden conditions to those exhibited by wild-caught toads from QLD and WA, we used a subset of

94 the data on wild toads (limited to individuals from those two regions). We used multiple regression with state and sex and their interaction as independent variables. As an additional, more formal test of the effect of rearing environment on locational differences in morphological sexual size dimorphism (SSD), we performed multiple regression combining data from wild-caught and common garden toads. The model included full factorial interactions among sex, state (WA versus QLD) and source (wild versus common garden).

SVL was included as a covariate in the model to control for body size relationships with limb lengths. A caveat for this analysis is that because we lacked data on relatedness among wild toads, we excluded familial effects from the model. Thus sibships among common garden toads were ignored.

To assess familial similarity in limb morphology in a formal quantitative genetics framework, we also ran an ‘animal model’ using ASREML software (VSN International,

Hemel Hempstead, UK: Wilson et al., 2010). When pedigree information is available (as is the case for our common garden offspring), animal models can be used to estimate the genetic underpinning of phenotypic variation (Wilson et al., 2010). Because most individuals were measured on more than one occasion, we were also able to estimate the ontogenetic repeatability of relative limb lengths. We incorporated offspring ID and parental ID as random effects in the animal model and included SVL as a covariate to correct for body size and age. Although our sample was adequate to detect heritabilities, it was too small to calculate genetic correlations among traits (Wilson et al., 2010). All other analyses were performed using JMP 11 software (SAS Institute, Cary, NC).

Results

Sexual Dimorphism Varies With Time Since Colonisation in Field-caught Toads

Relative limb lengths of wild-caught toads changed in different patterns with TSC in males versus females. Interactions between sex and TSC were highly significant both for arm length and leg length (Table 1 and Figure 1). Both sexes exhibited a ‘U’-shaped’ (curvilinear) pattern with TSC for both limbs (Figure 1); relative limb lengths were lowest in populations of intermediate TSC (11–20 years) than in either range-core or invasion-vanguard populations. Males had longer limbs than females in every population, but the magnitude of dimorphism was lower in recently invaded areas (i.e., Northern Territory and Western

Australia) than in areas that were colonised many decades ago (i.e., Hawai’i, Queensland).

The highest values for relative limb lengths in females were seen in invasion-front populations, whereas the highest values for males were seen in long-established populations.

Sexual Dimorphism Varies With Parental Location Among Common Garden Toads

Among the 184 captive-reared toads that reached maturity by the end of the study, the relative lengths of the humerus, foot, tibia and leg were affected by interactions between sex and the state of origin of their parents (QLD versus WA; Table 2 and Figure 2). These interactions broadly mirrored those seen in wild-caught toads from QLD versus WA (Table 3 and Figure 2). However, among wild toads, significant sex*state interactions were detected in all limb measures, whereas the same interactions were not statistically significant for some measures (hand, radioulna, arm and femur) among our smaller sample of common garden toads (Table 2). When we assessed the relative lengths of the humerus, foot, tibia and leg in an analysis that combined data from the 184 common garden toads with the 1274 wild toads from the source populations, it verified the same significant interaction effect of state and sex found when each group was analysed separately (Table 4). For all these measures, wild toads

96 had significantly longer relative limb lengths than common garden toads. However, although source (wild versus common garden) was a significant main effect in all analyses, it did not appear in any significant interactions (Table 4). Hence, wild-caught toads and those reared in captivity exhibited the same underlying pattern of decreased sexual dimorphism in relative limb lengths in invasive (WA) versus established (QLD) populations. This further supports the genetic basis of the patterns.

Estimates of Heritability

Our estimates of heritability for relative limb measures (Table 5) ranged from 0.02 ± 0.021

(femur) to 0.17 ± 0.043 (hand). These values suggest that most measures of limb morphology have a genetic component and hence are capable of responding to selection. The heritability estimates for femur and humerus, however, were low, and the confidence limits around them encompassed 0. These two traits may be less likely to respond to selection. We were able to measure limb components with high repeatability (Table 5). Tibia length was the most highly repeatable measurement (0.99), and hand was the least repeatable (0.80). As toads grew larger, relative lengths of their limbs were moderately stable; ontogenetic repeatability ranged from 0.19 (foot) to 0.39 (tibia).

Table 1. Effect of sex and time since colonisation (TSC) categories on relative arm and leg lengths of 3093 wild-caught cane toads from Hawai’i (N = 1018) and Australia (N = 2075).

Statistically significant values (P < 0.05) are highlighted in boldface font. SVL = snout-vent length.

Trait Variable df Mean Square F-value P-value Total arm length SVL 1 241558.3 19738.4 <0.0001 Sex 1 1973.7 161.3 <0.0001 TSC 4 1302.6 106.4 <0.0001 Sex*TSC 4 407.3 33.3 <0.0001 Error 3080 12.2 Total leg length SVL 1 704776.5 21233.9 <0.0001 Sex 1 2961.7 89.2 <0.0001 TSC 4 3534.1 106.5 <0.0001 Sex*TSC 4 760.6 22.9 <0.0001 Error 3080 33.2

Table 2. ANOVA results showing effects of Sex and State on limb sizes of 184 cane toads reared under common-garden conditions. 30 full-sib families produced offspring that reached maturity during the study and could be sexed based on secondary sexual characteristics. The parents of 17 families originated from long-established populations (Queensland) and the parents of 13 families originated from invasion-front populations (Western Australia). Only the final set of morphological measurements taken from each individual were included in this analysis. Snout-vent length (SVL) was included as a covariate in the model to adjust trait measures for body size. Family (nested within State) was included as a random effect in the model.

Trait SVL Sex State Sex * State F P F P F P F P Hand 260.8 <0.0001 5.1 0.0255 5.9 0.0225 1.5 0.2255 Radio-ulna 430.8 <0.0001 68.9 <0.0001 2.0 0.1647 0.4 0.5380 Humerus 466.3 <0.0001 94.5 <0.0001 1.0 0.3167 7.3 0.0075 Total arm 624.0 <0.0001 78.7 <0.0001 1.2 0.2804 1.1 0.2996 Foot 557.5 <0.0001 14.5 0.0002 0.4 0.5495 4.7 0.0315 Tibia 783.1 <0.0001 42.0 <0.0001 0.6 0.4666 4.8 0.0300 Femur 422.4 <0.0001 21.8 <0.0001 1.0 0.3316 3.6 0.0587 Total leg 717.1 <0.0001 29.2 <0.0001 0.7 0.4020 5.7 0.0177

Table 3. ANOVA results showing effects of Sex and State on limb sizes of 1274 wild-caught cane toads from long-established populations in Queensland (N = 467) and invasion-front populations in Western Australia (N = 807). SVL = snout-vent length.

SVL Sex State Sex * State Trait F P F P F P F P Hand 3082.7 <0.0001 32.1 <0.0001 39.1 <0.0001 26.1 <0.0001 Radio-ulna 4853.4 <0.0001 291.5 <0.0001 8.4 0.0037 61.6 <0.0001 Humerus 4607.6 <0.0001 413.4 <0.0001 1.6 0.2087 51.4 <0.0001

Total arm 6714.4 <0.0001 357.5 <0.0001 17.1 <0.0001 73.2 <0.0001 Foot 5127.2 <0.0001 90.6 <0.0001 24.8 <0.0001 42.5 <0.0001 Tibia 8189.1 <0.0001 260.7 <0.0001 11.1 0.0009 71.5 <0.0001

Femur 5025.2 <0.0001 119.5 <0.0001 12.4 0.0004 36.4 <0.0001 Total leg 7615.8 <0.0001 174.7 <0.0001 22.8 <0.0001 61.4 <0.0001

100

Table 4. Effects of Sex, State and Source (wild vs. common-garden) on limb sizes of 1274 wild-caught and 184 captive-reared toads from long-established populations in Queensland and invasion-front populations in Western Australia. The significant main effect of Source in all cases indicates that wild toads had relatively longer limbs than captive-reared toads. The significant shift in sexual dimorphism of limb lengths between States (indicated by significant Sex*State interactions) does not differ between wild-caught and common-garden toads (all three-way interactions NS). SVL = snout-vent length.

Trait Effect df F-value P-value

Total Leg Length SVL 1,1447 8339.8 <0.0001 State 1,1447 9.3 0.0023 Sex 1,1447 94.2 <0.0001 State*Sex 1,1447 21.9 <0.0001 Source 1,1447 182.3 <0.0001 State*Source 1,1447 0.23 0.6351 Sex*Source 1,1447 0.5 0.4884 State*Sex*Source 1,1447 0.4 0.5403 Humerus SVL 1,1449 5086.1 <0.0001 State 1,1449 3.5 0.0628 Sex 1,1449 243.8 <0.0001 State*Sex 1,1449 21.9 <0.0001 Source 1,1449 114.5 <0.0001 State*Source 1,1449 0.7 0.4204 Sex*Source 1,1449 2.3 0.1277 State*Sex*Source 1,1449 0.04 0.8440 Tibia SVL 1,1449 9008.1 <0.0001 State 1,1449 5.6 0.0180 Sex 1,1449 136.7 <0.0001 State*Sex 1,1449 23.4 <0.0001 Source 1,1449 167.1 <0.0001 State*Source 1,1449 0.01 0.9194 Sex*Source 1,1449 0.3 0.5648 State*Sex*Source 1,1449 0.7 0.3887 Foot SVL 1,1447 5697.3 <0.0001 State 1,1447 6.9 0.0087 Sex 1,1447 47.7 <0.0001 State*Sex 1,1447 14.7 0.0001 Source 1,1447 121.7 <0.0001 State*Source 1,1447 1.1 0.2939 Sex*Source 1,1447 0.2 0.6916 State*Sex*Source 1,1447 0.4 0.5483

101

Table 5. Estimates of heritability, ontogenetic repeatability (of measures made at different ages on the same animals over a long period), and measurement repeatability (of successive measures taken on the same animal over a brief period) of limb measurements of cane toads reared in a common-garden environment. Heritability and ontogenetic repeatability estimates were calculated from data on 550 individual toads (489 offspring, 61 parents). Measurement repeatability was calculated from a sample of 5 toads from which triplicate measures were made on a single occasion.

Ontogenetic Measurement Trait Heritability Repeatability Repeatability Hand 0.17 ± 0.043 0.34 ± 0.033 0.80 ± 0.203 Radioulna 0.17 ± 0.052 0.35 ± 0.033 0.91 ± 0.137 Humerus 0.03 ± 0.024 0.33 ± 0.032 0.89 ± 0.147 Total arm 0.13 ± 0.046 0.29 ± 0.033 0.92 ± 0.122 Foot 0.11 ± 0.034 0.19 ± 0.033 0.97 ± 0.092

Tibia 0.11 ± 0.044 0.39 ± 0.032 0.99 ± 0.025 Femur 0.02 ± 0.021 0.23 ± 0.033 0.93 ± 0.101 Total leg 0.10 ± 0.037 0.23 ± 0.033 0.98 ± 0.043

102

Figure 1. Comparisons of relative leg (upper left panel) and arm (lower left panel) lengths of

3093 wild-caught male and female cane toads from locations with different colonisation times. Males are represented by open symbols and dashed lines, females by closed symbols and solid lines. Relative limb length values are expressed as percentage of body length.

Sexual size dimorphism (SSD) indices for each limb (calculated as mean male value minus mean female value) are presented in the corresponding right-hand panels. Error bars represent one standard error from the mean.

103

Figure 2. Relative limb lengths of cane toads from long-colonised versus invasion-front populations, as a function of whether the toads were wild-caught (right-hand panels) or captive-raised (left-hand panels). The left-hand panels show statistically significant interactions between sex and state for limb measures of male (open symbols, dashed line) and female (closed symbols, solid line) cane toads reared in a common-garden environment. The parents of these toads originated either from long-established populations in Queensland

(QLD) or from invasive populations in Western Australia (WA). Significant interactions for the same limb measures seen in wild male and female toads from the same locations are depicted in the right-hand panels. In each case, the limb length is expressed as a percentage of toad snout-vent length.

104

Discussion

The cane toad invasion of Australia has been accompanied by rapid changes to skeletal morphology. Among wild toads, relative leg lengths (ancestrally high) have decreased in the course of the invasion across Australia, and then increased again in invasion-vanguard populations. In contrast, sexual dimorphism in relative limb lengths exhibits a simpler monotonic decline: toads in ancestral (range-core) populations are highly dimorphic, whereas toads in invasion-front populations show little sex-based divergence in limb lengths. Relative limb length exhibits significant heritability, and captive-raised toads show similar patterns of morphology as their wild-caught parental populations, suggesting that these morphological shifts may represent evolved changes rather than (or as well as) developmentally plastic responses to different environments.

First, what processes have driven the curvilinear pattern of changes in relative limb length during the toads’ Australian invasion? Annual rates of range expansion increased substantially over this period (from 10 to 15 km per annum to more than 60 km per annum;

Phillips et al., 2006), potentially placing major stresses on a body plan that is poorly suited to continuous long-distance travel (Brown et al., 2007; Shilton et al., 2008). Long limbs provide high propulsive power for leaping, an effective tactic to evade an oncoming predator (Gans &

Parsons, 1966; Emerson, 1978; Zug, 1985; Essner et al., 2010) but may be poorly suited to long periods of continuous slow dispersal over irregular terrain. Cane toads are capable of multiple locomotor modes, and small frequent hops (‘bounding’) may be more energetically efficient at traversing long distances than is a reliance on large single hops (‘leaping’: Reilly et al., 2015). In addition, shorter hops may reduce biomechanical stresses on the toad’s body.

Long legs in invasion-front cane toads are associated with a high incidence of spinal arthritis

(Brown et al., 2007; Shilton et al., 2008). Hence, natural selection in the course of the toad’s long march across tropical Australia may have favoured individuals with shorter-than-

105 average arms and legs that moved by bounding rather than by leaping. The arms play a major functional role in bounding (Akella & Gillis, 2011; Griep et al., 2013; Reilly et al., 2015), consistent with the shifts seen in both forelimbs and hindlimbs.

Why has this process reversed in populations close to the invasion front, with longer- limbed individuals (of both sexes) in these western sites? This reversal may be due to spatial sorting rather than natural selection. Even if selection favours animals with shorter limbs, the rate of dispersal is highest for long-legged toads (based on radio-tracking: Phillips et al.,

2006). Thus, alleles for longer legs accumulate in the vanguard of the invasion, regardless of whether or not they enhance fitness of their bearers (Shine et al., 2011). The result is that leg length decreases over the course of the invasion, but with a reversal close to the invasion front where spatial sorting overrides natural selection. Alternatively (or additionally), selective advantages that accrue to individuals in the invasion vanguard (more food, owing to lower densities of conspecifics) may favour maximal dispersal rates (and thus, longer legs) in this phase of the invasion. Faster dispersal would not confer the same fitness benefits in longer-colonised areas, because it would not enable individuals to reach low-density populations. Our data thus extend and clarify a previous report of longer hindlimbs in toads close to the invasion front (Phillips et al., 2006). Our extensive sampling reveals a more complex scenario, with invasion driving a reduction in relative limb length, but reversing to a rapid increase in leg (and arm) length near the invasion front.

In contrast to the curvilinear trends in relative limb lengths with TSC, the magnitude of sexual dimorphism in limb dimensions showed a rapid decline in populations colonised between 40 and 20 years ago (Figure 1). Consistent with their greater limb lengths, male cane toads can travel faster than females (C.M. Hudson 2017, unpublished data from locomotor trials; see Hudson et al., 2016b for a similar sex difference in agility). All else being equal then, the evolution of more rapid dispersal during the toad invasion would have resulted in

106 males substantially out-pacing females, thereby reaping the benefits of enhanced food availability at the invasion front (Brown et al., 2013), but at the cost of a highly skewed operational sex ratio (or in the extreme, a lack of females). The high degree of sex-based divergence in relative limb length in cane toads from range-core populations may result from sexual selection; previous studies have documented an association between limb muscle mass and reproductive success in male cane toads in the field (Lee, 2001; Lee & Corrales, 2002).

Hence, selection may have favoured larger and/or more muscular limbs in male toads than females in ancestral populations; this dimorphic condition is widespread among anurans in general, including bufonids (Wells, 2010). As soon as toads began dispersing westwards from

Queensland, however (and especially, when that rate of dispersal accelerated: Urban et al.,

2007), novel forms of selection could have come into play, reshaping the ancestral cane toad body plan. First, low-density populations may render male-male competition less important at the invasion front (unless that trend is opposed by a shift in the operational sex ratio) and second, limb lengths affect not only potential dispersal rates, but also the energy and wear- and-tear associated with long-distance travel (Brown et al., 2007). We cannot tease apart the relative importance of those two processes – sexual selection and natural selection – in driving cane toads towards sexual monomorphism, nor can we convincingly distinguish the impact of spatial sorting from selection. Nonetheless, our data strongly support the a priori prediction that a biological invasion can impose novel evolutionary forces that reduce the degree of sexual dimorphism in ancestral (range-core) populations.

We base our interpretative scenarios on adaptive mechanisms shifting morphological traits of toads over the course of their invasion. The similarity of traits between common garden-reared and wild toads and the non-zero heritability suggest that the trait changes have a genetic basis and are capable of responding to selection. Despite the plastic effects wrought by rearing differences (i.e., wild toads have significantly longer limbs than captive toads), the

107 shift in limb dimorphism between established and invasive populations is strongly evident.

Verification that the differences in traits have arisen through selection would require Qst –

Fst analysis to compare shifts in the quantitative traits to concurrent shifts in neutral traits, however (Alho et al., 2011).

The heritabilities we calculated for relative limb length – around 10% – are lower than reported for morphological traits in many other species of animals (Mousseau & Roff, 1987).

Clearly, that leaves room also for significant environmental influences. Future work could usefully explore the sensitivity of toad limb lengths to larval conditions (as in Alho et al.,

2011) and characterise the mating systems of toads at the invasion front compared with range-core populations. Regardless of uncertainty about causal mechanisms, however, our data document a substantial shift in morphology and sexual dimorphism within an invasive species, within the span of a single human lifetime.

108

Acknowledgements

We thank Georgia Ward-Fear, Matthew Greenlees, Martin Mayer, Michael Crossland, Jayna

DeVore, Samantha McCann, Simon Ducatez, Crystal Kelehear, Chalene Bezzina, Greg

Clarke, Maddie Sanders, Jodie Gruber, Damian Holden, Georgia Kosmala, Marta Vidal

García and Kimberley Lemmen for assistance with toad collection; Melanie Elphick for formatting the manuscript; and William Mautz for information and resources in Hawai’i. We also thank Juha Merilä for his suggestions to improve the manuscript.

Literature Cited

Akella, T., Gillis, G.B. 2011. Hopping isn't always about the legs: forelimb muscle activity

patterns during toad locomotion. J. Exp. Zool. A 315: 1–11.

Alford, R.A., Brown, G.P., Schwarzkopf, L., Phillips, B.L., Shine, R. 2009. Comparisons

through time and space suggest rapid evolution of dispersal behaviour in an invasive

species. Wildl. Res. 36: 23–28.

Alford, R.A., Cohen, M.P., Crossland, M.R., Hearnden, M.N., James, D., Schwarzkopf, L.

1995. Population Biology of Bufo marinus in northern Australia. Pages 173–181 in

Finlayson, M., ed., Wetland Research in the Wet-Dry Tropics of Australia. Office of the

Supervising Scientist Report 101, Canberra, Australia.

Alho, J.S., Herczeg, G., Laugen, A.T., Räsänen, K., Laurila, A., Merilä, J. 2011. Allen’s rule

revisited: quantitative genetics of extremity length in the common frog along a

latitudinal gradient. J. Evol. Biol. 24: 59–70.

Andersson, M. 1994. Sexual Selection. Princeton University Press, Princeton, New Jersey.

Arnqvist, G., Rowe, L. 2005. Sexual Conflict. Princeton University Press, Princeton, New

Jersey.

109

Berthouly-Salazar, C., van Rensburg, B.J., Le Roux, J.J., van Vuuren, B.J., Hui, C. 2012.

Spatial sorting drives morphological variation in the invasive bird, Acridotheris tristis.

PLoS One 7: e38145.

Bitton, P.P., Graham, B.A. 2015. Change in wing morphology of the European starling

during and after colonization of North America. J. Zool. 295: 254–260.

Bonduriansky, R., Maklakov, A., Zajitschek, F., Brooks, R. 2008. Sexual selection, sexual

conflict and the evolution of ageing and life span. Funct. Ecol. 22: 443–453.

Brown, G.P., Kelehear, C., Shine, R. 2013. The early toad gets the worm: cane toads at an

invasion front benefit from higher prey availability. J. Anim. Ecol. 82: 854–862.

Brown, G.P., Phillips, B.L., Dubey, S., Shine, R. 2015. Invader immunology: invasion history

alters immune system function in cane toads (Rhinella marina) in tropical Australia.

Ecol. Lett. 18: 57–65.

Brown, G.P., Phillips, B.L., Shine, R. 2014. The straight and narrow path: the evolution of

straight-line dispersal at a cane-toad invasion front. Proc. R. Soc. B 281: 20141385.

Brown, G.P., Shilton, C., Phillips, B.L., Shine, R. 2007. Invasion, stress, and spinal arthritis

in cane toads. Proc. Natl. Acad. Sci. USA 104: 17698–17700.

Carranza, J., Pérez-Barbería, F.J. 2007. Sexual selection and senescence: male size‐dimorphic

ungulates evolved relatively smaller molars than females. Am. Nat. 170: 370–380.

Chapman, T., Arnqvist, G., Bangham, J., Rowe, L. 2003. Sexual conflict. Trends Ecol. Evol.

18: 41–47.

Chuang, A., Peterson, C.R. 2015. Expanding population edges: theories, traits, and trade-offs.

Glob. Change Biol. 22: 494–512.

Cwynar, L.C., MacDonald, G.M. 1987. Geographical variation of lodgepole pine in relation

to population history. Am. Nat. 129: 463–469.

110

Ducatez, S., Crossland, M., Shine, R. 2016. Differences in developmental strategies between

long-settled and invasion-front populations of the cane toad in Australia. J. Evol. Biol.

29: 335–343.

Emerson, S.B. 1978. Allometry and jumping in frogs: helping the twain to meet. Evolution 32:

551–564.

Essner Jr. R.L., Suffian, D.J., Bishop, P.J., Reilly, S.M. 2010. Landing in basal frogs:

evidence of saltational patterns in the evolution of anuran locomotion.

Naturwissenschaften 97: 935–939.

Forsman, A., Merilä, J., Ebenhard, T. 2011. Phenotypic evolution of dispersal-enhancing

traits in insular voles. Proc. R. Soc. B 278: 225–232.

Freeland, W.J. 1986. Populations of cane toad Bufo marinus in relation to time since

colonization. Aust. Wildl. Res. 13: 321–330.

Gans, C., Parsons, T.S. 1966. On the origin of the jumping mechanism in frogs. Evolution 20:

92–99.

Griep, S., Schilling, N., Marshall, P., Amling, M., Hahme, L.M., Haas, A. 2013. Pectoral

girdle movements and the role of the glenohumeral joint during landing in the toad,

Rhinella marina (Linnaeus, 1758). Zoomorphology 132: 325–338.

Hagman, M., Hayes, R.A., Capon, R.J., Shine, R. 2009. Alarm cues experienced by cane toad

tadpoles affect post-metamorphic morphology and chemical defences. Funct. Ecol. 23:

126–132.

Hudson, C.M., Fu, J. 2013. Male-biased sexual size dimorphism, resource defense polygyny,

and multiple paternity in the Emei moustache toad (Leptobrachium boringii). PLoS

One 8: e67502.

Hudson, C.M., Phillips, B.L., Brown, G.P., Shine, R. 2015. Virgins in the vanguard: low

reproductive frequency in invasion front cane toads. Biol. J. Linn. Soc. 116: 743–747.

111

Hudson, C.M., McCurry, M.R., Lundgren, P., McHenry, C.R., Shine, R. 2016a. Constructing

an invasion machine: the rapid evolution of a dispersal-enhancing phenotype during the

cane toad invasion of Australia. PLoS One 11: e0156950.

Hudson, C.M., Brown, G.P., Shine, R. 2016b. Athletic anurans: the impact of morphology,

ecology and evolution on climbing ability in invasive cane toads. Biol. J. Linn. Soc. 116:

743–747.

Kokko, H., Rankin, D.J. 2006. Lonely hearts or sex in the city? Density-dependent effects in

mating systems. Phil. Trans. R. Soc. B 361: 319–334.

Lee, J.C. 2001. Evolution of secondary sexual dimorphism in the toad, Bufo marinus. Copeia

2001: 928–935.

Lee, J.C., Corrales, A.D. 2002. Sexual dimorphism in hind-limb muscle mass is associated

with male reproductive success in Bufo marinus. J. Herpetol. 36: 502–505.

Lindström, T., Brown, G.P., Sisson, S., Phillips, B.L., Shine, R. 2013. Rapid shifts in

dispersal behaviour on an expanding range edge. Proc. Natl. Acad. Sci. USA 110:

13452–13456.

Llewelyn, J., Phillips, B.L., Alford, R.A., Schwarzkopf, L., Shine, R. 2010. Locomotor

performance in an invasive species: cane toads from the invasion front have greater

endurance, but not speed, compared to conspecifics from a long colonised area.

Oecologia 162: 343–348.

Martin, O.Y., Hosken, D.J. 2003. The evolution of reproductive isolation through sexual

conflict. Nature 423: 979–982.

McCann, S., Greenlees, M.J., Newell, D., Shine, R. 2014. Rapid acclimation to cold allows

the cane toad to invade montane areas within its Australian range. Funct. Ecol. 28:

1166–1174.

112

Miller, T.E.X., Shaw, A.K., Inouye, B.D., Neubert, M.G. 2011. Sex-biased dispersal and the

speed of two-sex invasions. Am. Nat. 177: 549–561.

Moran, E.V., Alexander, J.M. 2014. Evolutionary responses to global change: lessons from

invasive species. Ecol. Lett. 17: 637–649.

Mousseau, T.A., Roff, D.A. 1987. Natural selection and the heritability of fitness components.

Heredity 59: 181–197.

Pearson, D., Shine, R., Williams, A. 2002. Geographic variation in sexual size dimorphism

within a single snake species (Morelia spilota, Pythonidae). Oecologia 131: 418–426.

Phillips, B.L., Brown, G.P., Greenlees, M., Webb, J.K., Shine, R. 2007. Rapid expansion of

the cane toad (Bufo marinus) invasion front in tropical Australia. Austral Ecol. 32:

169–176.

Phillips, B.L., Brown, G.P., Webb, J.K., Shine, R. 2006. Invasion and the evolution of speed

in toads. Nature 439: 803.

Phillips, B.L., Brown, G.P., Travis, J.M.J., Shine, R. 2008. Reid’s paradox revisited: the

evolution of dispersal kernels during range expansion. Am. Nat. 172: S34–S48.

Phillips, B.L., Brown, G.P., Shine, R. 2010a. Evolutionarily accelerated invasions: the rate of

dispersal evolves upwards during the range advance of cane toads. J. Evol. Biol. 23:

2595–2601.

Phillips, B.L., Brown, G.P., Shine, R. 2010b. Life-history evolution in range-shifting

populations. Ecology 91: 1617–1627.

Phillips, B.L., Kelehear, C., Pizzatto, L., Brown, G.P., Barton, D., Shine, R. 2010c. Parasites

and pathogens lag behind their host during periods of host range-advance. Ecology 91:

872–881.

113

Reilly, S.M., Montuelle, S.J., Schmidt, A., Naylor, E., Jorgensen, M.E., Halsey, L.G., Essner,

R.L. 2015. Conquering the world in leaps and bounds: hopping locomotion in toads is

actually bounding. Funct. Ecol. 29: 1308–1316.

Relyea, R.A. 2001. The lasting effects of adaptive plasticity: predator induced tadpoles

become long-legged frogs. Ecology 82: 1947–1955.

Relyea, R.A., Hoverman, J.T. 2003. The impact of larval predators and competitors on the

morphology and fitness of juvenile tree frogs. Oecologia 134: 596–604.

Rollins, L.A., Richardson, M.F., Shine, R. 2015. A genetic perspective on rapid evolution in

cane toads (Rhinella marina). Mol. Ecol. 24: 2264–2276.

Ronce, O., Clobert, J. 2012. Dispersal syndromes. Pages 119–138. in Clobert, J., Baguette,

M.G., Benton, T.G., Bullock, J.M., eds., Dispersal Ecology and Evolution. Oxford

University Press, Oxford, UK.

Schielzeth, H., Stoffel, M., Nakagawa, S. 2016. rptR: Repeatability Estimation For Gaussian

and Non-Gaussian Data. R package version 0.9.0.

Shilton, C.M., Brown, G.P., Benedict, S., Shine, R. 2008. Spinal arthropathy associated with

Ochrobactrum anthropi in free-ranging cane toads (Chaunus [Bufo] marinus) in

Australia. Vet. Pathol. 45: 85–94.

Shine, R. 1979. Sexual selection and sexual dimorphism in the Amphibia. Copeia 1979: 297–

306.

Shine, R., Brown, G.P., Phillips, B.L. 2011. An evolutionary process that assembles

phenotypes through space rather than through time. Proc. Natl. Acad. Sci. USA 108:

5708–5711.

Simmons, A.D., Thomas, C.D. 2004. Changes in dispersal during species’ range expansions.

Am. Nat. 164: 378–395.

114

Slip, D.J., Shine, R. 1988. The reproductive biology and mating system of diamond pythons,

Morelia spilota (Serpentes, Boidae). Herpetologica 44: 396–404.

Therry, L., Nilsson-Oertman, V., Bonte, D., Stoks, R. 2014. Rapid evolution of larval life

history, adult immune function and flight muscles in a poleward-moving damselfly. J.

Evol. Biol. 27: 141–152.

Tingley, R., Greenlees, M.J., Shine, R. 2012. Hydric balance and locomotor performance of

an anuran (Rhinella marina) invading the Australian arid zone. Oikos 121: 1959–1965.

Travers, L.M., Garcia-Gonzalez, F., Simmons, L.W. 2015. Live fast die young life history in

females: evolutionary trade-off between early life mating and lifespan in female

Drosophila melanogaster. Sci. Rep. 5: 15469.

Travis, J.M.J., Dytham, C. 2002. Dispersal evolution during invasions. Evol. Ecol. Res. 4:

1119–1129.

Urban, M.C., Phillips, B.L., Skelly, D.K., Shine, R. 2007. The cane toad’s (Chaunus [Bufo]

marinus) increasing ability to invade Australia is revealed by a dynamically updated

range model. Proc. R. Soc. B 274: 1413–1419.

Van Buskirk, J., Saxer, G. 2001. Delayed costs of an induced defense in tadpoles?

Morphology, hopping, and development rate at metamorphosis. Evolution 55: 821–829.

Wells, K.D. 2010. The Ecology and Behaviour of Amphibians. University of Chicago Press,

Chicago, IL.

Wilson, A.J., Reale, D., Clements, M.N., Morrissey, M.M., Postma, E., Walling, C.A., Kruuk,

L.E., Nussey, D.H. 2010. An ecologist’s guide to the animal model. J. Anim. Ecol. 79:

13–26.

Witjethunga, U., Greenlees, M., Shine, R. 2016. Living up to its name? The effect of salinity

on development, growth and phenotype of the “marine” toad (Rhinella marina). J.

Comp. Physiol. B 186: 205–213.

115

Zug, G.R. 1985. Anuran locomotion: fatigue and jumping performance. Herpetologica 41:

188–194.

116

Chapter 6

Effects of Toe-clipping on Growth, Body Condition, and Locomotion of

Cane Toads (Rhinella marina)

Manuscript submitted to Copeia as:

Hudson, C.M., Brown, G.P., Shine, R. Effects of toe-clipping on growth, body condition, and locomotion of cane toads (Rhinella marina).

117

Abstract

Toe-clipping is a standard technique for individually marking reptiles and amphibians, but concerns have been raised about the impact of the practice on animal welfare, survival, and behaviour. We used a long-term mark-recapture dataset to investigate the impact of toe removal on free-ranging adult cane toads (Rhinella marina). Records from 213 toads showed no impact of the number of toes removed on growth rates for mass or snout-urostyle length, nor any effect on body condition at recapture. Trials on a laboratory raceway revealed a short-term impact of toe-clipping on willingness to move (lower immediately post-clipping), but no other significant impacts on locomotion. In summary, toe-clipping appeared to have had minimal effects on cane toad locomotor ability, growth rate or body condition.

Key-words: Bufo marinus, digit removal, growth rates, mark-recapture, movement.

118

Introduction

Researchers working on amphibians use a variety of marking techniques to enable the repeated identification of individuals (Donnelly et al., 1994; Gibbons & Andrews, 2004;

Schmidt & Schwarzkopf, 2010). Ideally, such methods should have no impact on the welfare, behaviour, growth, or survival of the focal animal (Lemckert, 1996; Ferner, 2007; Guimarães et al., 2014). The sensitive skin and regenerative capabilities of amphibians preclude many potential methods of individual marking, leaving toe-clipping as one of the most commonly utilised methods (Ferner, 2007; Perry et al., 2011).

As well as enabling individual identification, toe-clipping can provide nonlethal tissue samples for genetic research (Gonser & Collura, 1996), detection of diseases such as chytridiomycosis (Boyle et al., 2004), and identifying exposure to toxic chemicals (Pfleeger et al., 2016). Removed toes can also be used to estimate individual ages through skeletochronology, providing information about population demography and individual longevity (Felton et al., 2006; Cheong et al., 2013; Buono et al., 2014). In comparison to alternative marking methods, toe-clipping provides a more reliable method for repeated identification than passive integrated transponder (PIT) tagging or visual implant elastomer

(VIE) tagging, because amphibians can expel foreign objects from their bodies (Tracy et al.,

2011; Brannelly et al., 2013, 2014; Kelehear et al., 2015). Thus, toe-clipping is in common use by amphibian biologists.

Toe-clipping has been criticised as invalid and unethical (May, 2004) based on a study that documented lower survival rates (as indicated by recapture probabilities) as a function of increasing numbers of toes removed (McCarthy & Parris, 2004). Other studies have documented infections and necrosis following clipping (Golay & Durrer, 1994). For genetic research, non-destructive sampling methods such as buccal, skin, and cloacal swabs offer viable alternatives (Pidancier et al., 2003; Broquet et al., 2007; Beja-Pereira et al., 2009),

119 although these procedures also suffer from shortcomings such as increased rates of non- target-DNA contamination (Müller et al., 2014). Several studies have tested the impact of toe-clipping (see Perry et al., 2011 for a review), or been written in defense of the practice

(Funk et al., 2005; Phillott et al., 2007, 2008), but the issue remains contentious. To obtain a broader view on the effects of this method, we need analyses both of long-term ecological correlates of toe removal (in the field), and of the short-term behavioural consequences of toe-clipping (in the laboratory).

Cane toads (Rhinella marina) are large, robust anurans that are among the world’s most successful invasive amphibians (Lowe et al., 2000; Lever, 2001), and have become a model system for studies on the ecological impacts of invasive species (Shine, 2010) and on the evolution of dispersal rates (Chuang & Peterson, 2016). Thus, it is important to assess the impact of marking techniques on the welfare and behaviour of this iconic species. Previous studies on stress responses to handling and toe-clipping in R. marina have produced conflicting results (Narayan et al., 2011, 2012; Fisher et al., 2013). Narayan et al. (2012) detected an increase in urinary corticosterone levels due to toe-clipping, whereas Fisher et al.

(2013) reported no significant difference in blood-plasma corticosterone levels between toads that were toe-clipped and those that were not. These studies on 'stress hormone' levels involved immediate responses to toe-clipping in animals held in captivity, a circumstance that is in itself stressful (deAssis et al., 2015). The effect of toe-clipping on the well-being of toads subsequently returned to the wild is unclear. During a 10-year mark-recapture study of invasive cane toads in the Northern Territory of Australia, we assessed the impact of toe- clipping on free-ranging toads by examining the relationship between the number of toes removed and the animal’s size and body condition at subsequent recapture events. We also explored potential short-term effects of toe-clipping on toad locomotor performance.

120

Materials and Methods

Mark-Recapture Study of Free-Ranging Toads

Beginning with the arrival of cane toads at our study site in tropical Australia, 60 km southeast of Darwin (Tropical Ecology Research Facility: 12.5673°S, 131.2996°E) in 2005, we have collected toads by hand, measured them with Vernier calipers (± 0.1 mm), and weighed them with spring scales (± 0.5 g). Following measurements, we toe-clipped each individual by removing the distal phalange from one to eight toes (no more than two digits per limb), giving each toad an easily identifiable combination (Donnelly et al., 1994). Toads were then released at their points of capture. Upon subsequent recaptures, toads were re- measured to collect data on body condition and growth. As an index of body condition at each capture time, we used residual scores from the general linear regression of ln body mass against ln snout-urostyle length (SUL).

Effect of Toe-Clipping on Locomotion

We collected 10 toads (mean SUL ± SE = 88.3 ± 1.3 mm) from a site in the Northern

Territory (Leaning Tree Lagoon: 12.7090°S, 131.3095°E). Toads were placed at the start of a

10-cm wide, 5-m long plastic-sided raceway, and encouraged to hop by prodding their urostyles with a blunt pole. We recorded the time, number of hops, and number of pokes for the toad to complete the racetrack. Each toad was trialled seven times between 20 August and

19 September 2014. The first trial was conducted in the field, followed by a laboratory trial after 24 h. We then toe-clipped each individual by removing the distal phalange from one to two toes in total on the hind feet, and trialled them again after 24 h. Following this, toads were housed outdoors in plastic containers (1 x 1 m) with a soil substrate, automated sprinklers, and artificial lights to attract insects to the enclosures. They were trialled in the same manner every seven days for the next four weeks. All trials were conducted between

121

1900 and 2400 h, indoors, at a controlled temperature (24°C). Ambient temperature during the field trial (at point of capture) was 17.9°C.

Statistical Analyses

We ran multiple regression models on cane toad growth rates (in mass, SUL and body condition) for individuals captured over periods of 30–90 days with initial size, month of capture, year of capture, and number of toes clipped as independent variables. We used one- way ANOVA (with trial as a factor) to compare the effect of toe-clipping on locomotor performance during repeated trials of recently captured toads. All statistical analyses were performed using JMP 11 software (SAS Institute, Cary, North Carolina).

Results

Mark-Recapture Study of Free-Ranging Toads

Multiple regression models on the growth rates of 213 recaptured toads (for mass, SUL and body condition: Table 1) showed no significant influence of the number of toes removed on individual growth (mass growth rate: # toes removed, N = 213, F1,212 = 0.63, P = 0.43; SUL growth rate: # toes removed, N = 213, F1,212 = 0.35, P = 0.56) or change in body condition (N

= 213, F1,212 = 1.54, P = 0.22). Month of capture, year of capture and initial measurements were all significant predictors of an individual’s growth rate (Table 1).

Effect of Toe-Clipping on Locomotion

A comparison between locomotor trials revealed no effect of toe-clipping (or captivity) on locomotion parameters (one-way ANOVA, N = 70, total hops, F1,69 = 1.64, P = 0.15; body lengths per hop, F1,69 = 2.17, P = 0.057; time to cover 5 m, F1,69 = 1.76, P = 0.12) with the exception of the total number of pokes required to complete the racetrack (one-way ANOVA,

122

N = 70, total pokes, F1,69 = 5.23, P = 0.0002). Immediately after toe-clipping, and one week following, toads required a significantly higher number of pokes in the urostyle to cover the 5 m distance (Figure 1).

123

Table 1. Multiple regression model results for the effect of number of toes removed, initial month of capture, year of capture, and measurements at first capture on growth rates (mass and snout-urostyle length [SUL]) of free-ranging cane toads. Significant values are highlighted in boldface.

Trait Variable df Test statistic P-value

Mass change/day # toes clipped 1 F = 0.63 0.43

Month of capture 11 F = 4.18 <0.0001

Year 9 F = 1.92 0.052

Mass (first capture) 1 F = 12.53 0.0005

SUL change/day # toes clipped 1 F = 0.35 0.56

Month of capture 11 F = 3.52 0.0002

Year 9 F = 4.03 <0.0001

SUL (first capture) 1 F = 68.60 <0.0001

Condition change/day # toes clipped 1 F = 1.54 0.22

Month of capture 11 F = 2.30 0.012

Year 9 F = 3.06 0.002

SUL (first capture) 1 F = 15.39 0.0001

124

Figure 1. Effect of toe-clipping on the number of pokes required to encourage a cane toad to hop the entire length of a 5-m racetrack; trials with asterisks are significantly different; error bars represent one standard error from the mean. The first three trials (field, lab, and toe-clip) were conducted on three consecutive nights, with the following trials each conducted one week apart.

125

Discussion

Soon after they were toe-clipped, cane toads were reluctant to move along a racetrack.

However, this behavioural impact disappeared within two weeks, and our field recaptures over a period of one to three months suggested that the number of toes removed had no significant impact on any of the viability-related measures for which we have data (rates of growth, and body condition). Overall, then, cane toads appear not to have been strongly affected by toe-clipping (or at least, by the number of toe-tips removed during that process).

Inevitably, there are caveats to our conclusions. First, individual toads at our study site are highly vagile (e.g. Lindström et al., 2013) and recaptures only occur within the 6- month dry season (May-November) each year (Brown et al., 2011). During the wet season toads disperse out of the study area, and we are unable to evaluate the impacts of toe-clipping on long-term survival. Given that toe-clipped animals grew normally and were in good body condition when recaptured, however, we doubt that toe removal has a survival cost. In support of that inference, we have often found wild cane toads with injuries (from predator attack, burns during wildfires, etc.) far more severe than those inflicted during toe-clipping.

Impacts of toe-clipping doubtless vary among anuran taxa (Perry et al., 2011). Cane toads are likely to be more resilient to toe-clipping than are many other anuran species, because these bufonids are large, robust, terrestrial, defended by toxic skin secretions, and capable of mounting a rapid and effective immune response to infection (Brown et al., 2011,

2015). A smaller, more fragile arboreal frog that relies upon its toes to cling to branches might well be far more sensitive to toe-tip removal than a cane toad. We encourage other fieldworkers with datasets similar to our own, to publish their analyses on this question. The potential impact of toe-tip removal on anuran viability is an important question both for animal welfare and for the validity of scientific conclusions based on data from toe-clipped

126 animals. Only with a broad taxonomic and geographic sampling of studies, will we be able to establish generalities about the impacts of marking methods on our anuran study animals.

127

Acknowledgements

We thank G. Kosmala for assistance with toad collection, M. Elphick for help with manuscript editing, and the Northern Territory Land Corporation for facilities. This study was funded by the Australian Research Council, and approved by the Animal Care and Ethics

Committee of the University of Sydney (Protocol 6705).

Literature Cited

Beja-Pereira, A., Oliveira, R., Alves, P.C., Schwartz, M.K., Luikart, G. 2009. Advancing

ecological understandings through technical transformations in non-invasive genetics.

Mol. Ecol. Res. 9: 1279–1301.

Boyle, D.G., Boyle, D.B., Olsen, V., Morgan, A., Hyatt, D. 2004. Rapid quantitative

detection of chytridiomycosis (Batrachochytrium dendrobatidis) in amphibian

samples using real-time Taqman PCR assay. Dis. Aquat. Organ. 60: 141–148.

Brannelly, L.A., Berger, L., Skerratt, L.F. 2014. Comparison of three widely used marking

techniques for adult anuran species Litoria verrauxii alpine. Herpetol. Conserv. Biol.

9: 428–435.

Brannelly, L.A., Chatfield, M.W.H., Richards-Zawacki, C.L. 2013. Visual implant elastomer

(VIE) tags are an unreliable method of identification in adult anurans. Herpetol. J. 23:

125–129.

Broquet, T., Berset-Braendli, L., Emaresi, G., Fumigalli, L. 2007. Buccal swabs allow

efficient and reliable microsatellite genotyping in amphibians. Conserv. Genet. 8:

509–511.

Brown, G.P., Kelehear, C., Shine, R. 2011. Effects of seasonal aridity on the ecology and

behaviour of invasive cane toads in the Australian wet-dry tropics. Funct. Ecol. 25:

1339–1347.

128

Brown, G.P., Shilton, C.M., Shine, R. 2011. Measuring amphibian immunocompetence:

validation of the phytohemaglutinin skin-swelling assay in the cane toad, Rhinella

marina. Methods Ecol. Evol. 2: 341–348.

Brown, G.P., Phillips, B.L., Dubey, S., Shine, R. 2015. Invader immunology: invasion history

alters immune system function in cane toads (Rhinella marina) in tropical Australia.

Ecol. Lett. 18: 57–65.

Buono, V., Guarino, F.M., Vignoli, L. 2014. Maximum body size and age distribution in the

Italian Stream Frog, Rana italica Dubois 1987 (Amphibia: Anura). Acta. Herpetol. 9:

231–235.

Cheong, S., Yoo, J., Park, S., Sung, H. 2013. Age estimation by skeletochronology and

advertisement call variation in the black-spotted pond frog (Rana nigromaculata).

Anim. Cells Syst. 17: 141–146.

Chuang, A., Peterson, C.R. 2015. Expanding population edges: theories, traits, and trade-offs.

Glob. Change Biol. 22: 494–512.

De Assis, V.R., Titon, S.C.M., Barsotti, A.M.G., Titon, Jr. B., Gomes, F.R. 2015. Effects of

acute restraint stress, prolonged captivity stress and transdermal corticosterone

application on immunocompetence and plasma levels of corticosterone on the Cururu

toad (Rhinella icterica). PLoS One 10: e0121005

Donnelly, M.A., Guyer, C., Juterbock, J.E., Alford, R.A. 1994. Techniques for marking

amphibians. Pages 277–284 in W.R. Heyer, M.A. Donnelly, R.W. McDiarmid, L.C.

Hayek, and M.S. Foster, eds., Measuring and Monitoring Biological Diversity:

Standard Methods for Amphibians. Smithsonian Institution Press, Washington, DC,

USA.

129

Felton, A.F., Alford, R.A., Felton, A.M., Schwarzkopf, L. 2006. Multiple mate choice criteria

and the importance of age for male mating success in the microhylid frog, Cophixalus

ornatus. Behav. Ecol. Sociobiol. 59: 786–795.

Ferner, J.W. 2007. A Review of Marking and Individual Recognition Techniques for

Amphibians and Reptiles. Herpetological Circular 35. Society for the Study of

Amphibians and Reptiles, Topeka, KS, USA.

Fisher, K., Guilfoyle, K., Hatch, K. 2013. Stress induced by toe-clipping in cane toads

(Rhinella marina). Copeia 2013: 539–542.

Funk, C.W., Donnelly, M.A., Lips, K.R. 2005. Alternative views of amphibian toe-clipping.

Nature 433: 193.

Gibbons, J.W., Andrews, K.M. 2004. PIT tagging: simple technology at its best. Bioscience

54: 447–454.

Golay, N., Durrer, H. 1994. Inflammation due to toe-clipping in Natterjack Toads (Bufo

calamita). Amphibia-Reptilia 15: 81–83.

Gonser, R.A., Collura, R.V. 1996. Waste not, want not: toe-clips as a source of DNA. J.

Herpetol. 30: 445–447.

Guimarães, M., Corrêa, D.T., Filho, S.S., Oliveira, T.A., Doherty, P.F., Sawaya, R.J. 2014.

One step forward: contrasting the effects of toe clipping and PIT tagging on frog

survival and recapture probability. Ecol. Evol. 4: 1480–1490.

Kelehear, C., Jones, H.I., Wood, B.A., Shine, R. 2015. Wild cane toads (Rhinella marina)

expel foreign matter from the coelom via the urinary bladder in response to internal

injury, endoparasites and disease. PLoS One 10: e0134036.

Lemckert, F. 1996. Effects of toe-clipping on the survival and behavior of the Australian frog

Crinia signifera. Amphibia-Reptilia 17: 287–290.

130

Lever, C. 2001. The Cane Toad: The History and Ecology of a Successful Colonist. Westbury

Academic and Scientific Publishing, Otley, West Yorkshire, UK.

Lindström, T., Brown, G.P., Sisson, S.A., Phillips, B.L., Shine, R. 2013. Rapid shifts in

dispersal behavior on an expanding range edge. Proc. Natl. Acad. Sci. USA 110:

13452–13456.

Lowe, S., Browne, M., Boudjelas, S., De Poorter, M. 2000. 100 of the World’s Most Invasive

Alien Species: A Selection from the Global Invasive Species Database. Invasive

Species Specialist Group, Auckland, New Zealand.

May, R.M. 2004. Ethics and amphibians. Nature 431: 403.

McCarthy, M.A., Parris, K.M. 2004. Clarifying the effect of toe clipping on frogs with

Bayesian statistics. J. Appl. Ecol. 41: 780–786.

Müller, A.S., Lenhardt, P.P., Theissinger, K. 2014. Pros and cons of external swabbing of

amphibians for genetic analyses. Eur. J. Wildl. Res. 59: 609–612.

Narayan, E.J., Molinia, F.C., Kindermann, C., Cockrem, J.F., Hero, J.M. 2011. Urinary

corticosterone responses to capture and toe-clipping in the cane toad (Rhinella

marina) indicate that toe-clipping is a stressor for amphibians. Gen. Comp.

Endocrinol. 174: 238–245.

Narayan, E.J., Hero, J.M., Cockrem, J.F. 2012. Inverse urinary corticosterone and

testosterone metabolite responses to different durations of restraint in the cane toad

(Rhinella marina). Gen. Comp. Endocrinol. 179: 345–349.

Perry, G., Wallace, M.C., Perry, D., Curzer, H., Muhlberger, P. 2011. Toe clipping of

amphibians and reptiles: science, ethics, and the law. J. Herpetol. 45: 547–555.

Pfleeger, A.Z., Eagles-Smith, C.A., Kowalski, B.M., Herring, G., Willacker, Jr. J.J., Jackson,

A.K., Pierce, J.R. 2016. From tails to toes: developing nonlethal tissue indicators of

mercury exposure in five amphibian species. Ecotoxicology 25: 574–583.

131

Phillott, A.D., Skerratt, L.F., McDonald, K.R., Lemckert, F.L., Hines, H.B., Clarke, J.M.,

Alford, R.A., Speare, R. 2007. Toe-clipping as an acceptable method of identifying

individual anurans in mark-recapture studies. Herpetol. Rev. 38: 305–308.

Phillott, A.D., Skerratt, L.F., McDonald, K.R., Lemckert, F.L., Hines, H.B., Clarke, J.M.,

Alford, R.A., Speare, R. 2008. Toe clipping of anurans for mark-recapture studies:

acceptable if justified. That’s what we said! Herpetol. Rev. 39: 149–150.

Pidancier, N., Miquel, C., Miaud, C. 2003. Buccal swabs as a non-destructive tissue sampling

method for DNA analysis in amphibians. Herpetol. J. 13: 175–178.

Schmidt, K., Schwarzkopf, L. 2010. Visible implant elastomer tagging and toe-clipping:

effects of marking on locomotor performance of frogs and skinks. Herpetol. J. 20:

99–105.

Shine, R. 2010. The ecological impact of invasive cane toads (Bufo marinus) in Australia. Q.

Rev. Biol. 85: 253–291.

Tracy, C.R., Christian, K.A., McArthur, L.J., Gienger, C.M. 2011. Removing the rubbish:

frogs eliminate foreign objects from the body cavity through the bladder. Biol. Lett. 7:

465–467.

132

Chapter 7

The Not-So-Great Escape: Evolutionary Shifts in Anti-predator Responses of Cane Toads (Rhinella marina)

Manuscript submitted to Behavioural Ecology and Sociobiology as:

Hudson, C.M., Brown, G.P., Shine, R. 2017. The not-so-great escape: evolutionary shifts in anti-predator responses of cane toads (Rhinella marina).

133

Abstract

A potential prey item’s response to encountering a predator depends on aspects of the predator (e.g. its locomotor capacity), the local environment (e.g. proximity to shelter) and the physiological state of the prey item; but in addition, anti-predator tactics differ consistently among individuals from different populations within wide-ranging species.

Using standardised trials, we tested responses of cane toads (Rhinella marina) to being placed on a laboratory runway and encouraged to flee. Overall, the toads least capable of rapid locomotion were the ones most likely to respond to simulated predator attack by exuding toxins rather than attempting to escape. A toad’s willingness to move down the runway, and its propensity to exude toxin from the parotoid glands rather than fleeing, were repeatable in successive trials, and depended on the animal’s (a) location of origin (specimens from

Australia were more willing to flee than were those from the native range [French Guiana] or

Hawai’i); (b) morphology (larger toads, and those with relatively longer legs, were more willing to flee); (c) previous experience (captive-raised toads were less willing to flee, and more willing to exude toxin); and (d) parentage (captive-raised offspring resembled their wild-caught parents both in propensity to run and in propensity to exude toxin). Thus, geographic divergence among cane toad populations in anti-predator responses reflects a complex combination of processes, including both developmental plasticity and heritability.

Key-words: anti-predator response, Bufo marinus, invasive species, locomotion, morphology,

Rhinella marina.

134

Introduction

When approached by a predator, animals use a variety of responses to avoid capture or evade detection. These include, but are not limited to: crypsis (Donnelly & Dill, 1984; Broom &

Ruxton, 2005; Caro, 2005; Barbosa et al., 2012), tonic immobility or death feigning

(Hoagland, 1928; Ewell et al., 1981; Arduino & Gould, 1984; Miyatake et al., 2004; Gregory et al., 2007; Durso & Mullin, 2013), flight (Hertz et al., 1982; Cooper & Frederick, 2007), warning calls (Seyfarth et al., 1980; Magrath et al., 2007; Natale et al., 2010), defensive or distractive displays (Brodie & Gibson, 1969; Greene, 1988; Caro, 2005; Hossie & Sherratt,

2013), discharge of foul or toxic compounds (Toledo & Jared, 1995; Williams et al., 2000;

Hopkins & Migabo, 2010; Medill et al., 2011; Toledo et al., 2011; Mailho-Fontana et al.,

2014), and combat (Caro, 2005; Emlen, 2008; Stankowich, 2010).

Such tactics may be contextual (Ydenberg & Dill, 1986; Lima & Dill, 1990), and depend on current environmental conditions (e.g. temperature: Cooper, 2000; Shine et al.,

2000; de Barros et al., 2009; habitat complexity: Pennings, 1990; Gifford et al., 2008), prior experience of the prey (Cox & Lima, 2006; Atkins et al., 2016), type of predator (Seyfarth et al., 1980; Dutour et al., 2016), and body condition (Amo et al., 2007), age (Hawlena et al.,

2006; Landová et al., 2013; Cooper, 2015; Putman et al., 2015), or reproductive status of the prey (Cushing, 1985; Burger et al., 1989; Shine et al., 2000; Brown & Shine 2004). However intrinsic factors such as genotype (Brodie, 1993), maternal effects (Shine & Downes, 1999;

Bestion et al., 2014), developmental temperature (particularly in ectotherms: Amiel & Shine,

2012; Amiel et al., 2014; Hagman et al., 2015) and morphology (Brodie, 1989; Eklöv &

Werner, 2000; de Barros et al., 2009; Mayer et al., 2016) can constrain the plasticity of anti- predator responses. Reflecting such intrinsic factors, anti-predator tactics may vary among populations of wide-ranging species, in response to variation in factors such as predator types and abundances (e.g. Shine et al., 2003).

135

Documenting impacts of predator attributes (e.g. size, deadliness, locomotor ability), local environmental conditions (e.g. proximity to shelter, habitat complexity) and prey physiological state (e.g. body temperature, reproductive status) on anti-predator tactics is relatively straightforward and amenable to experimental manipulation. The role of intrinsic factors is more difficult to explore. Significant repeatability of anti-predator responses among individuals testifies to the importance of such factors (e.g. Brodie et al., 1993; Brodie &

Russell, 1999; Edelaar et al., 2012;), but in order to clarify the processes at work, we need to conduct standardised trials (to eliminate variance induced by the conditions of the encounter) to ask:

(1) Do conspecific animals from different areas respond differently?

(2) Is an individual’s anti-predator behaviour associated with its size and morphology?

(3) Are anti-predator responses repeatable within individuals?

(4) Does developmental plasticity (e.g. rearing environment) affect anti-predator tactics? and

(5) Are anti-predator behaviours heritable?

To examine these issues, I conducted standardised trials to simulate predator attack on

>600 cane toads (Rhinella marina). This species is native to South and Central America, but has been translocated to many other countries (enabling us to compare behaviour of individuals from a range of locations differing in suites of predators); and reacts to predator attack either by fleeing or by exuding powerful defensive chemicals from its parotoid

(shoulder) glands (Hostetler & Cannon, 1974; Toledo et al., 1992; Toledo & Jared, 1995;

Almeida et al., 2007; Jared et al., 2009; Mailho-Fontana et al., 2014). An individual’s dependence on fleeing versus active defence (toxin exudation), or the combination of these two tactics, thus is straightforward to quantify.

136

Materials and Methods

Specimen Capture, Morphological Measurements, and Collection Sites

From October 2013 to August 2016, I collected cane toads from locations within Australia,

Hawai’i, and French Guiana (see Table 1 for details). All toads were collected by hand and I used Vernier callipers (± 0.1 mm) to measure the snout-vent length (SVL) and limb lengths of each toad, and spring scales (± 0.5 g) to measure total body mass. The hand, radioulna and humerus were measured for the forelimb (arm), whereas the femur, tibiofibula and foot were measured for the hindlimb (leg). Values for each component of a limb were added together to obtain measures of total forelimb length and total hindlimb length. A subgroup of adult toads from Queensland (QLD; N = 105) and Western Australia (WA; N = 159) were bred in captivity in order to conduct a common-garden breeding experiment. The resulting offspring were raised at a facility near Darwin (in the Northern Territory [NT]). All other toads were trialled within 48 h of capture. Toads were sexed by examining external morphological characteristics and vocalisations (see Hudson et al., 2016a for details). I considered toads to be immature if they measured <90 mm in SVL (Alford et al., 1995).

Common-Garden Toads

I used adult toads collected from the two extremes of the invaded Australian range to conduct a “common-garden” breeding experiment. These individuals were sourced from three long- established populations in northeastern Queensland (>70 years since colonisation: Townsville,

Innisfail, Tully) and four recently-colonised sites in northern Western Australia (<3 years since colonisation: El Questro, Purnululu, Wyndham, Oombulgurri; see Phillips et al., 2010;

Hudson et al., 2015; 2016a for details on spawning and rearing conditions). From these field- collected adults I produced 31 egg clutches (16 Queensland, 15 Western Australia), and obtained data on 489 captive-raised offspring (287 Queensland, 202 Western Australia). I

137 took measurements of the same morphological traits on these offspring as I did on wild toads.

In Sept-Oct 2015 I used 211 common-garden offspring in anti-predator trials (132

Queensland, 79 Western Australia) and conducted a second trial (to assess repeatability) on

177 of the same individuals (124 Queensland, 53 Western Australia) approximately four months later (Jan 2016). Mean SVL for common-garden offspring was 83.8 ± 0.62 mm during the first trial, and 93.0 ± 0.67 mm during the second trial.

Trials of Anti-Predator Responses

I tested the anti-predator responses of wild-caught adult cane toads (Queensland, N = 100;

Northern Territory, N = 30; Western Australia, N = 159; Hawai’i, N = 93; French Guiana, N =

49), and captive raised common-garden progeny (Queensland, N = 132; Western Australia, N

= 79) using standardised trials. Toads were placed at the start of a 2-m wide, 15-m long outdoor raceway and encouraged to hop by prodding their urostyle with a blunt pole. Toads that refused to hop after 10 consecutive pokes were considered to be exhausted or unwilling to move, and their trial was terminated. These individuals were excluded from the analysis of locomotion. Throughout the trials I recorded the time, number of hops, and number of pokes for the toad to complete each 5-m segment of the racetrack, as well as whether a toad exuded toxin from its skin during the trial. All trials were conducted between 1900 h and 2400 h, and

I recorded the average ambient temperature during the trial periods.

Statistical Analyses

I used a Principal Components Analysis (PCA) on locomotion parameters (total number of hops, total number of pokes, and time to move 15 m) to condense correlated data into a single variable representing locomotor style. The first Principal Component (hereafter dubbed

“Reluctance score”) represented 69.4% of the variation, with all parameters loading in the

138 same direction. Individuals with positive Reluctance scores were slower to flee, and slower at completing the raceway; negative values represent fast individuals, that required little encouragement to run. Following the PCA analysis, I used multivariate linear and logistic regressions to explore relationships among morphological, behavioural, and geographic parameters at the population level. Statistical analyses were performed using JMP 11 software (SAS Institute, Cary, NC).

To assess familial similarity in anti-predator behaviour in a formal quantitative genetics framework, I also ran an “animal model” using ASREML software (VSN International,

Hemel Hempstead, UK: Wilson et al., 2009). When pedigree information is available (as is the case for our common-garden offspring), animal models can be used to estimate the genetic underpinning of phenotypic variation (Wilson et al., 2010). Because many individuals were used in repeated locomotor trials, I was also able to estimate the repeatability of behavioural traits. I incorporated offspring ID and parental ID as random effects in the animal model and included SVL as a covariate to correct for body size and age. SVL was not included as a covariate for the logistic regression used to calculate heritability and repeatability of toxin exudation, because the model failed to converge when it was included.

Results

(a) Comparison among countries

Larger toads (greater SVLs) and those with relatively longer legs were less reluctant to run than were smaller, shorter-limbed conspecifics (Table 2), and toads from Australia were less reluctant to run than were French Guianan and Hawai’ian toads (Figure 1A).

139

(b) Comparison among Australian locations

Toads from the invasion-front (WA) were more reluctant to run than were toads from longer- colonised areas (QLD and NT; see Table 2, Figure 1B). As above, longer-legged toads were less reluctant to run, and toads were less reluctant to run when tested under warmer conditions.

A comparison of QLD and WA toads, but incorporating common-garden offspring (first trials only), produced similar effects to those above. Toads were less reluctant to run at higher temperatures, larger toads were less reluctant to run, and toads with relatively longer legs were less reluctant to run (Table 3). Toads collected near the invasion front (in WA) were more reluctant to run than were toads from long-colonised areas (QLD), and common-garden toads were more reluctant to run than were wild-caught toads (Figure 1C; the State*captivity interaction was non-significant).

Propensity to exude toxin

(a) Comparison among countries

Smaller toads and those with relatively short legs were more likely to exude toxin in response to simulated predation, but I found no significant geographic variation in this trait (Table 4).

Combining data for all toads, individuals that were unwilling or unable to complete the trial

(refused to move after 10 consecutive pokes) were more likely to exude toxin than were those that completed the 15-m racetrack (χ2 = 18.64, df = 1,611, P < 0.0001).

140

(b) Comparison among Australian states

When the analysis was restricted to data from Australia only, none of the independent variables explained significant variation in propensity to exude toxin (Table 4).

Common-garden toads were more likely to exude toxin during trials than were wild-caught toads (Table 3). Toads were more likely to exude toxin during trials at cooler temperatures, and toads with relatively small glands were more likely to exude toxin (Table 3). For common-garden toads (i.e. those raised in captivity), a greater proportion of individuals expelled toxin in the second trial than the first (χ2 = 8.37, df = 1,388, P one-tailed < 0.0003), and individuals that exuded toxin in the first trial were more likely to do so in the second (χ2 =

8.37, df = 1,388, P two-tailed < 0.0005).

Estimates of heritability

My estimates of heritability for anti-predator response (Table 5) ranged from 0.23 ± 0.081

(Time to run 15 m) to 0.47 ± 0.202 (Likelihood of exuding toxin). These values suggest that most measures of toad escape behaviour have a genetic component and hence are capable of responding to selection. Probability of toxin exudation was the most highly repeatable measurement (0.53 ± 0.087) and time to complete the 15-m racetrack was the least repeatable

(0.25 ± 0.075).

141

Table 1. Locations of capture for all wild toads.

Region Population #Male #Female Latitude Longitude

Queensland Innisfail 13 14 -17.524681 146.032329

Townsville 19 20 -19.257627 146.817871

Tully 20 14 -17.932869 145.923556

Northern Fogg Dam 15 15 -12.568032 131.309507

Territory

Western El Questro 26 19 -16.008438 127.979811

Australia Purnululu 20 22 -17.529752 128.400838

Oombulgurri 8 10 -15.180417 127.845039

Wyndham 31 23 -15.464803 128.100143

Hawai’i Richardson Park 8 4 19.736008 -155.013274

Wailoa River 6 6 19.720698 -155.07788

Panaewa Zoo 15 10 19.653754 -155.073765

Kona Life Care 15 12 19.571108 -155.957051

Mauna Lani G.C. 15 2 19.942773 -155.862629

French Matoury 10 6 4.8913537 -52.3338313

Guiana Remire-Montjoly 21 12 4.9170728 -52.2669545

142

Table 2. Effects of temperature, body size (snout-vent length [SVL]), relative leg length, parotoid gland size, body condition and location of origin on cane toad “reluctance to flee” scores. The first three results columns show results from analyses of the combined data set from all countries sampled (French Guiana, Hawai’i and Australia), whereas the next three columns show the results from analyses restricted to data from Australian toads only (from three Australian states: Queensland, Northern Territory and Western Australia). Residual scores are derived from general linear regressions of the trait versus ln-transformed snout- vent length. Significant values are highlighted in boldface.

Countries Australian States

Variable df F P df F P

Temperature 1 2.70 0.1011 1 12.29 0.0005

SVL 1 11.02 0.0010 1 0.40 0.5259

Residuals total leg length 1 9.06 0.0028 1 4.80 0.0293

Residuals gland circumference 1 1.13 0.2882 1 0.00 0.9878

Residuals log mass 1 0.53 0.4677 1 0.10 0.7550

Country / State 2 18.07 <0.0001 2 13.88 <0.0001

143

Table 3. Effects of temperature, body size (snout-vent length [SVL]), relative leg length, parotoid gland size, body condition, location of origin and captivity on cane toad “reluctance to flee” scores and probability of exuding toxin during the trial. Residual scores are derived from general linear regressions of the trait versus ln-transformed snout-vent length.

Significant values are highlighted in boldface. CG = common-garden, captive-raised.

Reluctance Probability of Exudation

2 Variable df F P df χ P

Temperature 1 22.00 <0.0001 1 7.79 0.0053

SVL 1 7.23 0.0074 1 2.59 0.1073

Residuals total leg length 1 12.63 0.0004 1 0.38 0.5400

Residuals gland circumference 1 0.91 0.3402 1 6.95 0.0084

Residuals log mass 1 1.68 0.1952 1 0.80 0.3711

State 1 9.97 0.0017 1 0.00 0.9656

Wild or CG 1 13.16 0.0003 1 14.54 0.0001

Wild or CG*State 1 1.33 0.2487 1 0.81 0.3683

144

Table 4. Effects of temperature, body size (snout-vent length [SVL]), relative leg length, parotoid gland size, body condition and location of origin on likelihood of toxin exudation.

The first three results columns show results from analyses of the combined data set from all countries sampled (French Guiana, Hawai’i and Australia), whereas the next three columns show the results from analyses restricted to data from Australian toads only (from three

Australian states: Queensland, Northern Territory and Western Australia). Residual scores are derived from general linear regressions of the trait versus ln-transformed snout-vent length.

Significant values are highlighted in boldface.

Countries Australian States

2 2 Variable df χ P df χ P

Temperature 1 3.59 0.0581 1 1.57 0.2095

SVL 1 4.10 0.0429 1 1.13 0.2886

Residuals total leg length 1 4.65 0.0310 1 0.09 0.7695

Residuals gland circumference 1 0.52 0.4692 1 0.04 0.8334

Residuals log mass 1 0.41 0.5206 1 0.19 0.6619

Country / State 2 3.29 0.1933 2 0.11 0.7445

145

Table 5. Estimates of heritability and repeatability (from multiple locomotor trials) of cane toads reared in a common-garden environment. Heritability and repeatability estimates were calculated from data on 273 individual toads (217 offspring, 56 parents).

Trait Heritability Repeatability

Reluctance 0.30 ± 0.089 0.38 ± 0.068

# Hops 0.26 ± 0.086 0.32 ± 0.073

# Pokes 0.32 ± 0.093 0.43 ± 0.064

Time to cover 15 m 0.23 ± 0.081 0.25 ± 0.075

Toxin Y/N 0.47 ± 0.202 0.53 ± 0.087

146

Figure 1. A comparison of “reluctance to flee” scores for wild-caught cane toads by country of origin (A), wild-caught Australian cane toads by state (B), and captive-raised F1 offspring

(solid line) by parental (dashed line) source population (C). Panels A and B display statistically significant relationships between population of origin and anti-predator response.

Panel C depicts a significant difference between toads from two Australian states in

“reluctance to flee”, and a non-significant interaction between captive-raised and wild-caught individuals.

147

Discussion

My simulated predation trials produced a diversity of responses from individual toads. Some refused to move (or moved only a very short distance) and instead exuded toxin, whereas others fled the entirety of the 15-m raceway with little stimulation. Overall, it appears that a cane toad’s anti-predator response represents a fight or flight trade-off (Cannon, 1929), explaining why reluctant runners are more likely to exude toxin. Toads that were very reluctant to move would begin exuding toxin soon after a trial commenced, whereas others did not cease running and begin exuding toxin until later in the trial, presumably because they were nearing exhaustion, or escalating their anti-predator response following continued harassment. By using a standardised design I eliminated many sources of variation (e.g. type of predator, stimulus, habitat type) and controlled for the (limited) range of thermal variation by including it in the analysis as a covariate. Thus, I can focus on residual sources of variation in response as a measure of anti-predator tactics.

From the captive-raised toads, I observed significant repeatability in both reluctance to flee, and likelihood of exuding toxin. Part of a toad’s anti-predator response is driven by biomechanical properties inherent to their morphology (Zug, 1972, 1978; Choi et al., 2003;

Gomes et al., 2009). The speed of a toad is enhanced by larger body size and relatively longer limbs, thus fleeing is a better option for them. Nonetheless, individual “personality” plays a role as well, as does learning. This is best demonstrated by the observation that captive toads were more likely to exude toxin during their second trial than the first, indicating an influence of prior experience on behaviour. Several studies have documented amphibians adjusting their anti-predator tactic in response to prior exposure to predators (Ferrari et al., 2008;

Teixeira & Young, 2014; Chivers et al., 2016). Therefore, at least some of the variation in anti-predator responses among populations may be the result of rearing environment and individual histories of encounters with predators.

148

There was a significant effect of location of origin on a toad’s anti-predator tactics.

Toads collected within Australia were more likely to flee than were native-range toads from

French Guiana, or toads from invasive populations in Hawai’i. Even within Australia, I recorded significant variation in response among populations, with invasion-front toads (from

WA) being more reluctant to run than were toads from longer-established populations (QLD and NT). Given the heritability for this trait, it is possible that shifts in anti-predator behaviour represent an evolved response. Below, I outline several scenarios in which these changes to anti-predator behaviour might be adaptive.

In the course of their Australian invasion, cane toads have come into contact with a variety of novel frog-eating predators that have no recent evolutionary history with bufonids.

Conversely, the invasive source population in Hawai’i is relatively predator-free, possibly explaining why Hawai’ian cane toads displayed poor running ability and high reluctance scores. Exhibiting a generalised flight response to any potential predator could be beneficial for survival of Australian cane toads, particularly when naïve predators have no information about prey toxicity. For an individual toad there is little benefit to being toxic if an encounter with a naïve predator leaves both prey and predator dead (e.g. snakes: Phillips & Shine,

2005a), therefore fleeing may be the better option. Previous research suggests that the size of the toxin-containing glands decreases with time since colonisation of an area (Phillips &

Shine, 2005b) possibly because there are fewer predators willing to consume toads (either via learning or local extinction), or because toxin is expensive to produce (Hettyey et al., 2014) and large quantities are no longer necessary for survival in a landscape with few resistant predators.

Why then, do we observe a difference in the anti-predator response between invasion- front and long-colonised populations? Toads at the invasion-front have been shown to be bolder and more exploratory when subjected to novel conditions in the laboratory (Gruber et

149 al., 2017), and respond less to stressful stimuli (e.g. produce less corticosterone in response to captivity: Brown et al., 2015; and adrenocorticotropic hormone injection; Hernández et al.,

2016). Therefore they may be less inclined to flee when disturbed by a novel predator. They also appear to be larger in body size, and more toxic (measured via gland size: Phillips &

Shine, 2005a) than their Australian conspecifics from long-established populations.

Alternatively, given the rapid timeframe of evolved changes within cane toads in Australia, some of the phenotypic divergences may be non-adaptive consequences of the rapid evolution of other traits (e.g. via pleiotropic effects). Viewed from this perspective, the observed shift in cane toad anti-predator response may actually be a side-effect of the evolution of distinctive behavioural, morphological, physiological phenotypes within

Australia. Because boldness and exploratory behaviour may enhance fitness in invasion-front toads (Gruber et al., 2017), these toads may also exhibit “bold” anti-predator responses when exposed to novel predators, even if such responses are maladaptive.

I found that a toad’s propensity to exude toxin was heritable, but I detected no significant geographic divergence in this trait. Instead, much of the variation in exudation likelihood (and reluctance to run) was elicited by rearing toads in captivity, rather than using wild-caught animals. This result suggests a strong role of developmental plasticity in cane toad anti-predator tactics. As previously documented in climbing ability of cane toads

(Hudson et al., 2016b) an animal’s prior experiences may influence its performance capacity.

Toads raised in our common-garden experiment were never confronted with predators (other than human researchers), and were unable to flee (as they were raised in 1 x 1 m containers), and hence may have relied on toxin exudation as their primary deterrent to predators.

The geographical divergence observed in cane toad anti-predator responses reflects a complex combination of processes, including both developmental plasticity and heritability.

Although I cannot definitively say whether or not the distinctive anti-predator tactics of cane

150 toads in Australia (compared to native-range and Hawai’ian conspecifics) reflect an adaptive response to novel predators, my data unequivocally demonstrate major shifts in behavioural attributes of toads in the course of the invasion process.

151

Acknowledgements

I thank Samantha McCann, Georgia Ward-Fear, Greg Clarke, Matthew Greenlees, Michael

Crossland, Simon Ducatez, Jayna DeVore, Crystal Kelehear, Martin Mayer, Maddie Sanders,

Jodie Gruber, Chalene Bezzina, Damian Holden, Georgia Kosmala, Marta Vidal García, and

Kimberley Lemmen for assistance with toad collection; Melanie Elphick for formatting the

manuscript, William Mautz for information and resources in Hawai’i, and Philippe Gaucher

for his assistance in co-ordinating field work in French Guiana. The study was funded by the

Australian Research Council, and conducted under permits from the University of Sydney

Animal Care and Ethics Committee.

Literature Cited

Alford, R.A., Cohen, M.P., Crossland, M.R., Hearnden, M.N., James, D., Schwarzkopf, L.

1995. Population biology of Bufo marinus in northern Australia. Pages 173–181 in M.

Finlayson, ed., Wetland Research in the Wet-Dry Tropics of Australia. Supervising

Scientist Report No. 101, Office of the Supervising Scientist, Canberra, ACT.

Almeida, P.G., Felsemburgh, F.A., Azevedo, R.A., Brito-Gitirana, L. 2007. Morphological

re-evaluation of the parotoid glands of Bufo ictericus (Amphibia, Anura, Bufonidae).

Contrib. Zool. 76:145–152.

Amiel, J.J., Shine, R. 2012. Hotter nests produce smarter young lizards. Biol. Lett. 8: 372–

374.

Amiel, J.J., Lindström, T., Shine, R. 2014. Egg incubation effects generate positive

correlations between size: speed and learning ability in young lizards. Anim. Cogn.

17: 337–347.

Amo, L., Lopez, P., Martın, J. 2007. Refuge use: a conflict between avoiding predation and

losing mass in lizards. Physiol. Behav. 90: 334–343.

152

Arduino, P.J., Gould, J.L. 1984. Is tonic immobility adaptive? Anim. Behav. 32: 921–923.

Atkins, R., Blumstein, D.T., Moseby, K.E., West, R., Hyatt, M., Letnic, M. 2016. Deep

evolutionary experience explains mammalian responses to predators. Behav. Ecol.

Sociobiol. 70: 1755–1763.

Barbosa, A., Allen, J.J., Mäthger, L.M., Hanlon, R.T. 2012. Cuttlefish use visual cues to

determine arm posture for camouflage. Proc. R. Soc. B 279: 84–90.

Bestion, E., Teyssier, A., Aubret, F., Clobert, J., Cote, J. 2014. Maternal exposure to predator

scents: offspring phenotypic adjustment and dispersal. Proc. R. Soc. B 281: 20140701.

Broom, M., Ruxton, G.D. 2005. You can run – or you can hide: optimal strategies for cryptic

prey against pursuit predators. Behav. Ecol. 16: 534–540.

Brodie Jr., E.D., Gibson, L.S. 1969. Defensive behavior and skin glands of the Northwestern

Salamander, Amybstoma gracile. Herpetologica 25: 187–194.

Brodie III, E.D. 1989. Genetic covariances between morphology and anti-predator behavior

in natural populations of the garter snake Thamnophis ordinoides. Nature 342: 542–

543.

Brodie III, E.D. 1993. Homogeneity of the genetic variance-covariance matrix for anti-

predator traits in two natural populations of the garter snake Thamnophis ordinoides.

Evolution 47: 844–854.

Brodie, E.D., Russell, N.H. 1999. The consistency of individual differences in behaviour:

temperature effects on antipredator behaviour in garter snakes. Anim. Behav. 57: 445–

451.

Brown, G.P., Shine, R. 2004. Effects of reproduction on the anti-predator tactics of snakes

(Tropidonophis mairii, Colubridae). Behav. Ecol. Sociobiol. 56: 257–262.

153

Brown, G.P., Kelehear, C., Shilton, C.M., Phillips, B.L., Shine, R. 2015. Stress and immunity

at the invasion front: a comparison across cane toad (Rhinella marina) populations.

Biol. J. Linn. Soc. 116: 748–760.

Burger, J., Gochfield, M., Saliva, J.E., Gochfield, D., Gochfield, D., Morlaes, H. 1989.

Antipredator behavior in nesting zenaida doves (Zenaida aurita): parental investment

or offspring vulnerability. Behaviour 111: 129–143.

Cannon, W.B. 1929. Bodily Changes in Pain, Hunger, Fear, and Rage. Appleton-Century-

Crofts, New York, USA.

Caro, T. 2005. Anti-predator Defenses in Birds and Mammals. University of Chicago Press,

Chicago, IL, USA.

Chivers, D.P., Mitchell, M.D., Lucon-Xiccato, T., Brown, G.E., Ferrari, M.C.O. 2016.

Background risk influences learning but not generalization of predators. Anim. Behav.

121: 185–189.

Choi, I., Shim, J.H., Ricklefs, R.E. 2003. Morphometric relationships of take-off speed in

anuran amphibians. J. Exp. Zool. 299: 99–102.

Cooper Jr., W.E. 2000. Effect of temperature on escape behaviour by an ectothermic

vertebrate, the keeled earless lizard (Holbrookia propinqua). Behaviour 137: 1299–

1315.

Cooper, W.E.J., Frederick, W.G. 2007. Optimal flight initiation distance. J. Theor. Biol. 244:

59–67.

Cooper, W. 2015. Age affects escape behavior by the zebra-tailed lizard (Callisaurus

draconoides) more strongly than in other lizards. Amphibia-Reptilia 36: 37–44.

Cox, J.G., Lima, S.L. 2006. Naiveté and an aquatic-terrestrial dichotomy in the effects of

introduced predators. Trends Ecol. Evol. 21: 674–680.

154

Cushing, B.S. 1985. Estrous mice and vulnerability to weasel predation. Ecology 66: 1976–

1978. de Barros, F.C., de Carvalho, J.E., Abe, A.S., Kohlsdorf, T. 2009. Fight versus flight: the

interaction of temperature and body size determines antipredator behaviour in tegu

lizards. Anim. Behav. 79: 83–88.

Donnelly, W.A., Dill, L.M. 1984. Evidence for crypsis in coho salmon, Oncorhynchus

kisutch (Walbaum), parr: substrate colour preference and achromatic reflectance. J.

Fish Biol. 25: 183–195.

Durso, A.M., Mullin, S.J. 2013. Intrinsic and extrinsic factors influence expression of

defensive behaviour in plains hog-nosed snakes (Heterodon nasicus). Ethology 120:

140–148.

Dutour, M., Lena, J.P., Lengagne, T. 2016. Mobbing behaviour varies according to predator

dangerousness and occurrence. Anim. Behav. 119: 119–124.

Edelaar, P., Serrano, D., Carrete, M., Blas, J., Potti, J., Tella, J.L. 2012. Tonic immobility is a

measure of boldness toward predators: an application of Bayesian structural equation

modeling. Behav. Ecol. 23: 619–626.

Eklöv, P., Werner, E.E. 2000. Multiple predator effects on size-dependent behaviour and

mortality of two species of anuran larvae. Oikos 88: 250–258.

Emlen, D.J. 2008. The evolution of animal weapons. Annu. Rev. Ecol. Evol. Syst. 39: 387–

413.

Ewell, A.H., Cullen, J.M., Woodruff, M.L. 1981. Tonic immobility as a predator-defense in

the rabbit (Oryctolagus cuniculus). Behav. Neural Biol. 31: 483–489.

Ferrari, M.C.O., Messier, F., Chivers, D.P. 2008. Larval amphibians learn to match

antipredator response to temporal patterns of risk. Behav. Ecol. 19: 980–983.

155

Gifford, M.E., Mahler, D.L., Herrel, A. 2008. The evolution of locomotor morphology,

performance, and anti-predator behavior among populations of Leiocephalus lizards

from the Dominican Republic. Biol. J. Linn. Soc. 93: 445–456.

Gomes, F.R., Rezende, E.L., Grizante, M.B., Navas, C.A. 2009. The evolution of jumping

performance in anurans: morphological correlates and ecological implications. J. Evol.

Biol. 22: 1088–1097.

Greene, H.W. 1988. Anti-predator mechanisms in reptiles. Pages 1–152 in C. Gans, R.B.

Huey, eds., Biology of the Reptilia, Ecology B. Defense and Life History, Volume 16.

Alan R. Liss Inc., New York, USA..

Gregory, P.T., Isaac, L.A., Griffiths, R.A. 2007. Death feigning by grass snakes (Natrix

natrix) in response to handling by human predators. J. Comp. Psychol. 121: 123–129.

Gruber, J., Brown, G., Whiting, M.J., Shine, R. 2017. Geographic divergence in dispersal-

related behaviour in cane toads from range-front versus range-core populations in

Australia. Behav. Ecol. Sociobiol. 71: 38.

Hagman, M., Löwenborg, K., Shine, R. 2015. Determinants of anti-predator tactics in

hatchling grass snakes (Natrix natrix). Behav. Process. 113: 60–65.

Hawlena, D., Boochnik, R., Abramsky, Z., Bouskila, A. 2006. Blue tail and striped body:

why do lizards change their infant costume when growing up? Behav. Ecol. 17: 889–

896.

Hernández, S.E., Sernia, C., Bradley, A.J. 2016. Adrenocortical function in cane toads from

different environments. Comp. Biochem. Physiol. A 195: 65–72.

Hertz, P.E., Huey, R.B., Nevo, E. 1982. Fight versus flight: body temperature influences

defensive responses of lizards. Anim. Behav. 30: 676–679.

Hettyey, A., Tóth, Z., Van Buskirk, J. 2014. Inducible chemical defences in animals. Oikos

123: 1025–1028.

156

Hoagland, H. 1928. On the mechanism of tonic immobility in vertebrates. J. Gen. Physiol.

11: 715–741.

Hopkins, G.R., Migabo, S.W. 2010. Antipredator skin sectretions of the long-toed salamander

(Ambystoma macrodactylum) in its northern range. J. Herpetol. 44: 627–633.

Hossie, T.J., Sherratt, T.N. 2013. Defensive posture and eyespots deter avian predators from

attacking caterpillar models. Anim. Behav. 86: 383–389.

Hostetler, J.R., Cannon, M.S. 1974. The anatomy of the parotoid gland in Bufonidae with

some histochemical findings. I. Bufo marinus. J. Morphol. 142: 225–239.

Hudson, C.M., Phillips, B.L., Brown, G.P., Shine, R. 2015. Virgins in the vanguard: low

reproductive frequency in invasion front cane toads. Biol. J. Linn. Soc. 116: 743–747.

Hudson, C.M., Brown, G.P., Shine, R. 2016a. It’s lonely at the front: contrasting evolutionary

trajectories in male and female invaders. Roy. Soc. Open Sci. 3: 160687.

Hudson, C.M., Brown, G.P., Shine, R. 2016b. Athletic anurans: the impact of morphology,

ecology and evolution on climbing ability in invasive cane toads. Biol. J. Linn. Soc.

116: 743–747.

Jared, C., Antoniazzi, M.M., Jordão, A.E.C., Silva, J.R., Greven, H., Rodrigues, M.T. 2009.

Parotoid macroglands in toad (Rhinella jimi): their structure and functioning in

passive defense. Toxicon 54: 197–207.

Landová, E., Jančúchová-Lásková, J., Musilová, V., Kadochová, S., Frynta, D. 2013.

Ontogenetic switch between alternative antipredatory strategies in the leopard gecko

(Eublepharis macularius): defensive threat versus escape. Behav. Ecol. Sociobiol. 67:

1113–1122.

Lima, S.L., Dill, L.M. 1990. Behavioural decisions made under the risk of predation: a review

and prospectus. Can. J. Zool. 68: 619–640.

157

Magrath, R.D., Pitcher, B.J., Gardner, J.L. 2007. A mutual understanding? Interspecific

responses by birds to each other’s aerial alarm calls. Behav. Ecol. 18: 944–951.

Mailho-Fontana, P.L., Antoniazzi, M.M., Toledo, L.F., Verdade, V.K., Sciani, J.M., Barbaro,

K.C., Pimenta, D.C., Rodrigues, M.T., Jared, C. 2014. Passive and active defense in

toads: the parotoid macroglands in Rhinella marina and Rhaebo guttatus. J. Exp. Zool.

321: 65–77.

Mayer, M., Shine, R., Brown, G.P. 2016. Bigger babies are bolder: effects of body size on

personality of hatchling snakes. Behaviour 153: 313–323.

Medill, S.A., Renard, A., Lariviére, S. 2011.Ontogeny of antipredator behavior in striped

skunks (Mephitis mephitis). Ethol. Ecol. Evol. 23: 41–48.

Miyatake, T., Katayama, K., Takeda, Y., Nakashima, A., Sugita, A., Mizumoto, M. 2004. Is

death-feigning adaptive? Heritable variation in fitness difference of death-feigning

behavior. Proc. R. Soc. B 271: 2293–2296.

Natale, G.S., Alcalde, L., Herrera, R., Cajade, R., Schafer, E.F., Marangoni, F., Trudeau, V.L.

2010. Underwater acoustic communication in the macrophagic carnivorous larvae of

Ceratophrys ornata. Acta. Zool. 92: 46–53.

Pennings, S.C. 1990. Predator-prey interactions in opisthobranch gastropods: effects of prey

body size and habitat complexity. Mar. Ecol. Prog. Ser. 62: 95–101.

Phillips, B.L., Shine, R. 2005a. Adapting to an invasive species: toxic cane toads induce

morphological change in Australian snakes. Proc. Natl. Acad. Sci. USA 101: 17150–

17155.

Phillips, B.L., Shine, R. 2005b. The morphology, and hence impact, of an invasive species

(the cane toad, Bufo marinus) changes with time since colonization. Anim. Conserv. 8:

407–413.

158

Phillips, B.L., Kelehear, C., Pizzatto, L., Brown, G.P., Barton, D., Shine, R. 2010. Parasites

and pathogens lag behind their host during periods of host range-advance. Ecology

91: 872–881.

Putman, B.J., Coss, R.G., Clark, R.W. 2015. The ontogeny of antipredator behaviour: age

differences in California ground squirrels (Otospermophilus beecheyi) at multiple

stages of rattlesnake encounters. Behav. Ecol. Sociobiol. 69: 1447–1457.

Seyfarth, R.M., Cheney, D.L., Marler, P. 1980. Monkey responses to three different alarm

calls: evidence of predator classification and semantic communication. Science 210:

801–803.

Shine, R., Downes, S.J. 1999. Can pregnant lizards adjust their offspring phenotypes to

environmental conditions? Oecologia 119: 1–8.

Shine, R., Olsson, M., Lemaster, M.P., Moore, I.T., Mason, R.T. 2000. Effects of sex, body

size, temperature, and location on the anti-predator tactics of free-ranging garter

snakes (Thamnophis sirtalis, Colubridae). Behav. Ecol. 11: 239–245.

Shine, R., Phillips, B., Waye, H., Mason, R.T. 2003. Small-scale geographic variation in

antipredator tactics of garter snakes. Herpetologica 59: 333–339.

Stankowich, T. 2010. Risk-taking in self-defense. Pages 79–86 in M.D. Breed, J. Moore, eds.,

Encyclopedia of Animal Behavior Volume 3. Academic Press, Oxford, UK.

Teixeira, B., Young, J. 2014. Can captive-bred American bullfrogs learn to avoid a model

avian predator? Acta. Ethol. 17: 15–22.

Toledo, R.C., Jared, C., Brunner Jr., A. 1992. Morphology of the large granular alveoli of

toad (Bufo ictericus) parotoid glands before and after compression. Toxicon 30: 745–

753.

Toledo, R.C., Jared, C. 1995. Cutaneous granular glands and amphibian venoms. Comp.

Biochem. Physiol. 111: 1–29.

159

Toledo, L.F., Sazima, I., Haddad, C.F.B. 2011. Behavioural defences of anurans: an overview.

Ethol. Ecol. Evol. 23: 1–25.

Williams, C.R., Brodie Jr., E.D., Tyler, M.J., Walker, S.J. 2000. Antipredator mechanisms of

Australian frogs. J. Herpetol. 34: 431–443.

Wilson, A.J., Reale, D., Clements, M.N., Morrissey, M.M., Postma, E., Walling, C.A., Kruuk,

L.E., Nussey, D.H. 2010 An ecologist’s guide to the animal model. J. Anim. Ecol. 79:

13–26.

Ydenberg, R.C., Dill, L.M. 1986. The economics of fleeing from predators. Adv. Stud. Behav.

16: 229–249.

Zug, G.R. 1972. Anuran locomotion: structure and function. 1. Preliminary observations on

the relation between jumping and osteometrics of appendicular and postaxial skeleton.

Copeia 1972: 613–624.

Zug, G.R. 1978. Anuran locomotion – structure and function, 2: jumping performance of

semiaquatic, terrestrial, and arboreal frogs. Smithson. Contrib. Zool. 276: 1–31.

160

Appendix A

First report of exotic ticks (Amblyomma rotundatum) parasitizing invasive cane toads (Rhinella marina) on the island of Hawai’i.

Manuscript published as:

Kelehear, C., Hudson, C.M., Mertins, J.W., Shine, R. 2017. First report of exotic ticks

(Amblyomma rotundatum) parasitizing invasive cane toads (Rhinella marina) on the island of

Hawai’i. Ticks and Tick-borne Diseases 8: 330–333. http://dx.doi.org/10.1016/j.ttbdis.2016.12.010

161

Abstract

Our surveys of 1401 invasive cane toads (Rhinella marina) from the Hawai’ian islands of

Hawai‘i, O‘ahu, and Maui revealed the presence of an exotic tick, Amblyomma rotundatum.

Immature and adult female ticks infested three wild adult toads at a single site in the vicinity of a zoo south of Hilo, Island of Hawai‘i, Hawai‘i, USA. We found no tick-infested toads on

O‘ahu or Maui. This tick infests cane toads in their native Neotropical range, but it was excluded from Hawai‘i when the original founder toads were introduced over 80 years ago.

The circumstances of our discovery suggest that A. rotundatum was independently and belatedly introduced to Hawai‘i with imported zoo animals, and Hawai‘i now joins Florida as the second U.S. state where this tick is established.

Key-words: Amblyomma rotundatum, amphibian, Anuran, Bufonidae, ectoparasite, Ixodidaea.

162

Introduction

Ever increasing levels of international travel and trade correspond to unprecedented rates of exotic species introductions (Costello & McAusland, 2003). Ticks are particularly adept at stowing away and entering new habitats on their introduced hosts (Keirans & Durden, 2001;

Barré & Uilenberg, 2010; Burridge, 2011). Although ticks are generally visible to the naked eye during quarantine inspections, their larval stages can be rather small, and some tick individuals may attach to hosts in sheltered anatomical sites (e.g. within nostrils or ear canals), where they are not easily dislodged or detected (Reeves et al., 2006). Moreover, ticks often remain attached to their hosts for sufficient periods of time (e.g. weeks or months) to endure conventional transit periods for live animals from one locality to another (Barré & Uilenberg,

2010; Luz et al., 2013).

The cane toad (Rhinella marina, formerly Bufo marinus) is a large bufonid anuran native to the Rio Grande Valley, southern Texas, USA, southwards to the southern Amazon

Basin in Brazil, but it also has been translocated to many other localities worldwide in attempts to control insect pests (Lever, 2001). Under either natural or captive conditions, R. marina is a known host for nine species of ixodid ticks and one argasid species (Burridge,

2011; Bermúdez et al., 2013), but of these, only Amblyomma dissimile and A. rotundatum are frequent and typical parasites (Guglielmone & Nava, 2010). All others seem to be aberrant, occasional, or opportunistic parasites. Tick infestations are common on cane toads throughout their native range (Lampo & Bayliss, 1996) and in some introduced populations in the

Caribbean (Newstead, 1909; Kohls, 1969; Drake et al., 2014) and Florida, USA (Oliver et al.,

1993; Goddard et al., 2015). Yet, ticks are absent from cane toads in other introduced populations (e.g. Bermuda, Australia, Hawai‘i), presumably because importers removed the ticks from the toads before their release in new localities (e.g. a radiogram was sent in 1932

163 by entomologist Cyril Pemberton to the receivers of the toads in Hawai’i, specifically to warn them to inspect the animals for ticks upon arrival: Turvey, 2013).

In 1932, 149 cane toads arrived on the Island of O‘ahu from an introduced source population in Puerto Rico, and the offspring (>600,000 toads) of those Hawai’ian immigrants were released on sugar plantations in the larger Hawaiian islands (Hawai‘i, Maui, O‘ahu,

Kaua‘i, Moloka‘i) between 1933 and 1935 (Easteal, 1981).

Materials and Methods

During 11–17 January 2015 and 3 June–29 July 2015, we inspected cane toads for tick infestations on three islands in the state of Hawai‘i. We hand-collected cane toads at night and visually inspected their skins for ticks. After inspections, the toads were euthanised by refrigeration, followed by freezing (Shine et al., 2015). We examined 693 toads from 11 populations on Hawai‘i, 189 toads from nine populations on Maui, and 519 toads from 14 populations on O‘ahu. We removed all ticks from infested toads, preserved these parasites in

70% ethanol, and identified them at the USDA National Veterinary Services Laboratories

(NVSL), Ames, Iowa. The identification process used morphological criteria, published reference materials (Keirans & Durden, 1998; Voltzit, 2007; Martins et al., 2010; Guzmán-

Cornejo et al., 2011), and comparison with archived reference specimens from toad hosts.

Voucher tick specimens are retained in the NVSL parasitology reference collection

(Accession No. 16-022027, Case No. T16-1180).

164

Results

We found ticks on toads at a single site south of Hilo (19.653754°N, 155.073765°W; Figure

1) on the Island of Hawai‘i on 15 July 2015; three (two females and one male) of the 47 toads

(6.38%) surveyed at that site were tick-infested. We identified these ticks as 2 females, 7 nymphs, and 1 larva of A. rotundatum. Most immatures (Figure 2A) and both females (Figure

2B and C) were partially engorged to different degrees. We found no ticks on any cane toad at all other sites in the state of Hawai‘i (Figure 1).

165

Figure 1. Map of sampling locations (open circles). (A) State of Hawai‘i, (B) Island of O‘ahu

(where we found no ticks on 519 toads surveyed across 14 populations), (C) Islandof Maui

(where we found no ticks on 189 toads surveyed across nine populations), and (D) Island of

Hawai‘i (where we surveyed 693 toads across 11 populations and found ticks on three toads from one of those populations; four-pointed star).

166

Figure 2. In situ photographs of ticks parasitising cane toads in Hawai‘i. (A) An Ambly- omma rotundatum nymph on the dorsum of a toad, (B) a greatly engorged female A. rotundatum on the side of a toad, and (C) a partially engorged female A. rotundatum on the ventral surface of a toad.

167

Discussion

Published records document only 14 tick species previously collected in the Hawai’ian

Islands (Joyce, 1980; Goff, 1987; Keirans & Durden, 2001); four of these were closely associated with resident seabirds, and the remainder were either known or suspected to have been introduced by human aegis. The Amblyomma ticks discovered in the present study probably fit into the latter category and were confined to a single site, infested only three toads, and consisted of 10 total individuals. These infestations consisted entirely of A. rotundatum, a widespread Neotropical tick native to southern Mexico, Central America, northern South America, and many Caribbean islands (Guglielmone et al., 2003), and introduced and established in southern Florida, USA (Oliver et al., 1993).

The list of hosts reportedly used by A. rotundatum is long and diverse (Guglielmone

& Nava, 2010; Scott & Durden, 2015), but the majority are reptiles and amphibians; rare single records from one bird and several mammal hosts are probably anomalies. Indeed, based upon criteria proposed by Guglielmone and Nava (2010), anuran amphibians are the usual natural hosts for all feeding stages of A. rotundatum, with Boa constrictor as a second- most-likely host, and bufonids overwhelmingly the most used hosts overall.

The single site where we found ticks was Pana’ewa Rainforest Zoo and Gardens, a small public zoo that exhibits a diverse array of non-native vertebrates. Zoo-keepers informed us (pers. comm.) that they sometimes find ticks on their Asian forest tortoises (Manouria emys phayrei), originally imported from Texas in 2000, and on green iguanas (Iguana iguana). A subsequent follow-up examination of resident reptiles at the zoo in September

2016 found small numbers of nymphal A. rotundatum present on an Asian forest tortoise and a red-footed tortoise (Chelonoidis carbonaria). Both wild and captive green iguanas (Dantas-

Torres et al., 2008; Guglielmone & Nava, 2010) and captive forest tortoises (USDA unpubl. data) are previously known hosts for A. rotundatum, and either could be the local introduction

168 source. The tortoises and iguanas are housed at the Hawai’ian zoo in large outdoor enclosures that are surrounded by fencing and vegetation. Cane toads are common in and around the exhibit areas (C.M.H. pers. obs.), allowing ample opportunity for the ticks to spread from the captive zoo animals to the wild cane toads. Three-host ticks like A. rotundatum require the availability of adequate and appropriate separate individual hosts for each of their three active life stages, and according to Guglielmone and Nava (2010), only B. constrictor and various anurans are known to serve A. rotundatum in this way. We found specimens of all three life stages feeding on toads, and it seems likely that the only way that populations of this tick can survive, reproduce, and persist in Hawai’i will be in association with resident introduced cane toads. Adult and nymphal ticks can feed and survive on other hosts, including green iguanas, and that is probably how A. rotundatum arrived in the zoo. However, any local reproduction that has taken place since that event has probably involved, at least in part, infestations of local cohabiting cane toads.

Encouragingly, the tick population seems not to have spread widely from the point of origin (all infested toads were collected within the grounds of the zoo), so an active control program could eliminate the ticks before they become more widely established. Such an effort would be wise because the ubiquity of cane toads on the island of Hawai‘i, and their competency as hosts for A. rotundatum, could facilitate the spread of these ticks across the entire island and, potentially, to new hosts (Corn et al., 2011; Kelehear et al., 2013).

Amblyomma rotundatum is associated with certain potentially adverse effects in its known hosts. In bufonid hosts, lesions form at tick attachment sites (Jackowska, 1972; Luz et al., 2013). In South America, intensities of tick infestations were negatively related to cane toad body condition (Lampo & Bayliss, 1996). In other taxa, effects can be even more pronounced; for example, a black racer (Coluber constrictor priapus) was paralysed by an attached A. rotundatum in Florida, USA (Hanson et al., 2007). This tick also is a vector of

169

Hemolivia stellata, a haemogregarine blood parasite of cane toads and other ectotherms (Petit et al., 1990; Boulard et al., 2001; Lainson et al., 2007), and is known to harbor Rickettsia bellii (Erster et al., 2015).

Conclusions

Herein, we report ticks of the genus Amblyomma on wild animals in the state of Hawai‘i for the first time (Garrett & Haramoto, 1967; Goff, 1987; Leong & Grace, 2008). The only previous reported occurrences of Amblyomma ticks in Hawai’i concerned four species

(Amblyomma americanum, A. dissimile, A. nuttalli, and A. sparsum) intercepted a total of eight times between 1964 and 1997 on imported hosts (Joyce, 1965, 1969, 1980; Keirans &

Durden, 2001; USDA unpub. data). None of these ticks became established in Hawai’i.

Amblyomma rotundatum may spread relatively easily in Hawai’i because of the ubiquity of invasive cane toads and because it is parthenogenetic, meaning a single tick could found a new population anywhere suitable hosts occur (Keirans & Oliver, 1993; Labruna et al., 2005).

Ticks are frequently introduced by humans to previously unreachable localities with imported animals. These exotic tick species usually are concentrated at various ports of entry, quarantine centres, airports, commercial breeding facilities, or zoos (Burridge et al., 2000;

Keirans & Durden, 2001; Burridge & Simmons, 2003; Kenny et al., 2004; Reeves et al.,

2006; Burridge, 2011; Mihalca, 2015). Our records of A. rotundatum from Hawai‘i provide another example of accidental long-distance translocation of ticks with exhibited animal hosts.

However, our documented tick introduction is unusual in that it initially went undetected, and the immigrant ticks were able to transfer to, and infest, a local wild host species. Moreover, this phenomenon is unique in that – after >80 years of forced separation – the infested “new” toad host is the tick’s most-used typical host in their shared ancestral home range.

Amblyomma rotundatum now joins two other immigrant tick species closely tied to

170 introduced hosts in Hawai’i; Otobius megnini and Rhipicephalus sanguineus were long ago accidentally introduced and established by humans with their typical hosts, domestic cattle and dogs, respectively (Bonnett, 1948). Finally, Hawai‘i now joins Florida as one of only two

U.S. states with established wild populations of A. rotundatum.

171

Acknowledgements

We thank Pam Mizuno for information, site access, and subsequent collection of ticks from captive tortoises. We thank Travis Heskett for instigating and expediting follow-up collection and shipment of ticks from zoo tortoises. Greg Clarke, Sam McCann, Jayna DeVore and

Chalene Bezzina assisted with toad collection. C. Kelehear was supported by the Global

Invasions Research Co-ordination Network, a George E. Burch Fellowship, and a National

Geographic Research and Exploration Grant #9945-16. R. Shine was supported by the

Australian Research Council.

Literature Cited

Barré, N., Uilenberg, G. 2010. Spread of parasites transported with their hosts: case study of

two species of cattle tick. Rev. Sci. Tech. Off. Int. Epiz. 29: 149–160.

Bermúdez, S.E.C., Miranda, R.J., Kadoch, N.S. 2013. Reporte de larvas de Ornithodoros

puertoricensis Fox 1947 (Ixodida: Argasidae) parasitando a Rhinella marina (L.

1758) (Anura: Bufonidae) en David Chiriquí, Panamá. Puente Biológico 5: 81–85.

Bonnett, D.D. 1948. Certain aspects of medical entomology in Hawaii. Proc. Hawaii

Entomol. Soc. 13: 225–233.

Boulard, Y., Paperna, I., Petit, G., Landau, I. 2001. Ultrastructure of developmental stages of

Hemolivia stellata (Apicomplexa: Haemogregarinidae) in the cane toad Bufo marinus

and in its vector tick Amblyomma rotondatum. Parasitol. Res. 87: 598–604.

Burridge, M.J., Simmons, L.A. 2003. Exotic ticks introduced into the United States on

imported reptiles from 1962 to 2001 and their potential roles in international

dissemination of diseases. Vet. Parasitol. 113: 289–320.

Burridge, M.J., Simmons, L-A., Allan, S.A. 2000. Introduction of potential heartwater

vectors and other exotic ticks into Florida on imported reptiles. J. Parasitol. 86: 700–

172

704.

Burridge, M.J. 2011. Non-native and Invasive Ticks: Threats to Human and Animal Health in

the United States. University Press of Florida, Gainesville, Florida.

Corn, J.L., Mertins, J.W., Hanson, B., Snow, S. 2011. First reports of ectoparasites collected

from wild-caught exotic reptiles in Florida. J. Med. Entomol. 48: 94–100.

Costello, C., McAusland, C. 2003. Protectionism, trade, and measures of damage from exotic

species introductions. Am. J. Agric. Econ. 85: 964–975.

Dantas-Torres, F., Oliveira-Filho, E.F., Soares, F.A.M., Souza, B.O.F., Valença, R.B.P., Sá,

F.B. 2008. Ticks infesting amphibians and reptiles in Pernambuco, Northeastern

Brazil. Rev. Bras. Parasitol. Vet. 17: 218–221.

Drake, M.C., Zieger, U., Groszkowski, A., Gallardo, B., Sages, P., Reavis, R., Faircloth, L.,

Jacobson, K., Lonce, N., Pinckney, R., Cole, R.A. 2014. Survey of helminths,

ectoparasites, and chytrid fungus of an introduced population of cane toads, Rhinella

marina (Anura: Bufonidae), from Grenada, West Indies. J. Parasitol. 100: 608–615.

Easteal, S. 1981. The history of introductions of Bufo marinus (Amphibia: Anura); a natural

experiment in evolution. Biol. J. Linn. Soc. 16: 93–113.

Erster, O., Roth, A., Avni, Z., King, R., Shkap, V. 2015. Molecular detection of Rickettsia

bellii in Amblyomma rotundatum from imported red-footed tortoise (Chelonoides

carbonaria). Ticks Tick-Borne Dis. 6: 473–477.

Garrett, L.E., Haramoto, F.H. 1967. A catalog of Hawaiian Acarina. Proc. Hawaii Entomol.

Soc. 19: 381–414.

Goddard, J., Ward, C., Dagg, J., Mendonça, M., Robbins, R.G. 2015. New records for

Amblyomma rotundatum Koch (Acari: Ixodidae) from the cane toad, Rhinella marina

L. in Florida, with notes on identification of the nymphs. Syst. Appl. Acarol. 20: 465–

467.

173

Goff, M.L. 1987. A catalog of Acari of the Hawaiian islands. Hawaii Inst. Trop. Agric. Hum.

Resour. Res. Ext. Ser. 75: 1–75.

Guglielmone, A.A., Nava, S. 2010. Hosts of Amblyomma dissimile Koch, 1844 and

Amblyomma rotundatum Koch, 1844 (Acari: Ixodidae). Zootaxa 2541: 27–49.

Guglielmone, A.A., Estrada-Peña, A., Keirans, J.E., Robbins, R.G. 2003. Ticks (Acari:

Ixodida) of the neotropical zoogeographic region. A special publication sponsored by

the International Consortium on Ticks and Tick-Borne Diseases, Atalanta, Houton,

The Netherlands.

Guzmán-Cornejo, C., Robbins, R.G., Guglielmone, A.A., Montiel-Parra, G., Pérez, T.M.

2011. The Amblyomma (Acari: Ixodida: Ixodidae) of Mexico: identification keys,

distribution and hosts. Zootaxa 2998: 16–38.

Hanson, B.A., Frank, P.A., Mertins, J.W., Corn, J.L. 2007. Tick paralysis of a snake caused

by Amblyomma rotundatum (Acari:Ixodidae). J. Med. Entomol. 44: 155–157.

Jackowska, S. 1972. Lesions produced by ticks, Amblyomma dissimile, in Bufo marinus toads

from the Dominican Republic. Am. Zool. 12: 731.

Joyce, C.R. 1965. Notes and exhibitions. Amblyomma americana (L.). Proc. Hawaii

Entomol. Soc. 19: 22–23.

Joyce, C.R. 1969. Notes and exhibitions. Amblyomma americanum (L.). Proc. Hawaii

Entomol. Soc. 20: 276.

Joyce, C.R. 1980. Notes and exhibitions. Amblyomma sparsum Neumann. Proc. Hawaii

Entomol. Soc. 23: 180.

Keirans, J.E., Durden, L.A. 1998. Illustrated key to nymphs of the tick genus Amblyomma

(Acari: Ixodidae) found in the United States. J. Med. Entomol. 35: 489–495.

Keirans, J.E., Durden, L.A. 2001. Invasion: exotic ticks (Acari: Argasidae, Ixodidae)

imported into the United States: a review and new records. J. Med. Entomol. 38: 850–

174

861.

Keirans, J.E., Oliver Jr., J.H. 1993. First description of the male and redescription of the

immature stages of Amblyomma rotundatum (Acari: Ixodidae) a recently discovered

tick in the U.S.A. J. Parasitol. 79: 860–865.

Kelehear, C., Brown, G.P., Shine, R. 2013. Invasive parasites in multiple invasive hosts: the

arrival of a new host revives a stalled prior parasite invasion. Oikos 122: 1317–1324.

Kenny, M.J., Shaw, S.E., Hillyard, P.D., Forbes, A.B. 2004. Ectoparasite and haemoparasite

risks associated with imported exotic reptiles. Vet. Rec. 154: 434–435.

Kohls, G.M. 1969. New records of ticks from the Lesser Antilles. Stud. Fauna. Curaçao

Caribbean Isl. 28: 126–134.

Labruna, M.B., Terrassini, F.A., Camargo, L.M. 2005. First report of the male of

Amblyomma rotundatum (Acari: Ixodidae) from a field-collected host. J. Med.

Entomol. 42: 945–947.

Lainson, R., De Souza, M.C., Franco, C.M. 2007. Natural and experimental infection of the

lizard Ameiva ameiva with Hemolivia stellata (Adeleina: Haemogregarinidae) of the

toad Bufo marinus. Parasite 14: 323–328.

Lampo, M., Bayliss, P. 1996. The impact of ticks on Bufo marinus from native habitats.

Parasitology 113: 199–206.

Leong, M.K.H., Grace, J.K. 2008. Occurrence and distribution of mites and ticks (Acari) of

public health importance on the Island of Oahu. Proc. Hawaii Entomol. Soc. 40: 19–

31.

Lever, C. 2001. The Cane Toad: The History and Ecology of a Successful Colonist.

Westbury Academic and Scientific Publishing, Otley, West Yorkshire.

Luz, H.R., Faccini, J.L.H., Pires, M.S., da Silva, H.R., Barros-Battesti, D.M. 2013. Life cycle

and behavior of Amblyomma rotundatum (Acari: Ixodidae) under laboratory

175

conditions and remarks on parasitism of toads in Brazil. Exp. Appl. Acarol. 60: 55–62.

Martins, T.F., Onofrio, V.C., Barros-Battesti, D.M., Labruna, M.B. 2010. Nymphs of the

genus Amblyomma (Acari: Ixodidae) of Brazil: descriptions, redescriptions, and

identification key. Ticks Tick-Borne Dis. 1: 75–99.

Mihalca, A.D. 2015. Ticks imported to Europe with exotic reptiles. Vet. Parasitol. 213: 67–

71.

Newstead, R. 1909. Ticks and other blood-sucking Arthropoda [in Jamaica]. Ann. Trop. Med.

Parasitol. 3: 421–469.

Oliver Jr., J.H., Hayes, M.P., Keirans, J.E., Lavender, D.R.1993. Establishment of the foreign

parthenogenetic tick Amblyomma rotundatum (Acari: Ixodidae) in Florida. J.

Parasitol. 79: 786–790.

Petit, G., Landau, I., Baccam, D., Lainson, R. 1990. Description et cycle biologique

d’Hemolivia stellata n. g. n. sp., hémogrégarine de crapauds brésiliens. Ann.

Parasitol. Hum. Comp. 65: 3–15.

Reeves, W.K., Durden, L.A., Dasch, G.A. 2006. A spotted fever group Rickettsia from an

exotic tick species, Amblyomma exornatum (Acari: Ixodidae), in a reptile breeding

facility in the United States. J. Med. Entomol. 43: 1099–1101.

Scott, J.D., Durden, L.A. 2015. First record of Amblyomma rotundatum tick (Acari: Ixodidae)

parasitizing a bird collected in Canada. Syst. Appl. Acarol. 20: 155–161.

Shine, R., Amiel, J., Munn, A.J., Stewart, M., Vyssotski, A.L., Lesku, J.A. 2015. Is cooling

then freezing a humane way to kill amphibians and reptiles? Biol. Open 4: 760–763.

Turvey, N. 2013. Cane Toads: A Tale of Sugar, Politics and Flawed Science. Sydney

University Press, Sydney, NSW.

Voltzit, O.V. 2007. A review of neotropical Amblyomma species (Acari: Ixodidae). Acarina

15: 3–134.

176

Appendix B

Sex and weaponry: the distribution of toxin-storage glands on the bodies of male and female cane toads (Rhinella marina)

Manuscript submitted to Ecology and Evolution as:

Chen, W., Hudson, C.M., DeVore, J.L., Shine, R. Sex and weaponry: the distribution of toxin-storage glands on the bodies of male and female cane toads (Rhinella marina).

177

Abstract

The parotoid macroglands of bufonid anurans store (and can expel) large volumes of toxic secretions, and have attracted detailed research. However, toxins also are stored in smaller glands that are distributed on the limbs and dorsal surface of the body. Female and male cane toads (Rhinella marina) differ in the location of toxin-storage glands and the extent of glandular material. Female toads store a larger proportion of their toxins in the parotoids than males, as well as (to a lesser extent) in smaller glands on the forelimbs. Males have smaller and more elongate parotoids than females, but glands cover more of the skin surface on their limbs (especially hindlimbs) and dorsal surface. The delay to toxin exudation in response to electro-stimulation varied among glands in various parts of the body, and did so differently in males than in females. The spatial distribution of toxin glands differs between the sexes even in toads that have been raised under standardised conditions in captivity; hence, the sexual dimorphism is due to heritable factors rather than developmentally plastic responses to ecological (e.g. habitat, predation risk) differences between the sexes. The selective advantages of this sexual dimorphism remain unclear. A priori, we might expect to see toxin widely dispersed across any part of the body likely to be contacted by a predator; and a wide distribution also would be expected if the gland secretions have other (e.g. male-male rivalry) functions. Why, then, is toxin concentrated in the parotoids, especially in female toads? That concentration may enhance the effectiveness of frontal displays to deter predation; and also, may facilitate the transfer of stored toxins to eggs.

Key-words: antipredator, bufadienalide, Bufo marinus, bufotoxin, chemical defence, sexual dimorphism.

178

Introduction

Although they share many genes in common, males and females within a population often differ in a wide range of phenotypic traits. The most easily-explained divergences involve characteristics with a direct role in reproductive biology, such as gonads, accessory glands, and sex-specific ornaments or weaponry (e.g. antlers in male deer: Clutton-Brock, 1991;

Andersson, 1994). However, the sexes also differ in a wide range of traits with no obvious link to reproduction. For example, males and females sometimes forage on different kinds of prey, or feed in different places using different tactics; and such ecological divergence can result in the evolution of sexual dimorphism in size or shape of the trophic apparatus (Shine,

1989; Neuhaus & Ruckstuhl, 2005; Fairbairn et al., 2007). Even the sensory modalities used to detect prey (e.g. Vincent et al., 2005) and cognitive abilities of the sexes (Carazo et al.,

2014) are subject to differential selective pressures.

Morphological divergence between the sexes is widespread in anuran amphibians, and often involves attributes such as mean adult body sizes and relative limb lengths (reflecting sexual selection and fecundity selection: Shine, 1979; Lee, 2001; Kupfer, 2007; Wells, 2010).

Direct physical combat between males has favoured the evolution of sex-specific weaponry such as pseudo-fangs and spines (Kastsikaros & Shine, 1997; Emerson, 2001; Tsuji & Matsui,

2002; Hudson et al., 2011) and increased musculature of the forelegs (Navas & James, 2007).

To our knowledge, though, antipredator tactics generally are similar between males and female anurans within the same population. Species (and even populations) differ considerably in traits such as the possession of defensive chemicals (Brodie et al., 2002), and colour polymorphism in many tropical anurans may have evolved in antipredator contexts

(see Wells, 2010 for a review). Nonetheless, we are unaware of any examples whereby conspecific male and female amphibians differ strongly in morphological traits that function

179 to deter predation. The closest example may be the shift away from crypsis towards bright colouration by male anurans at the peak of breeding activity (Doucet & Mennill, 2009).

Given that general lack of sexual dimorphism in antipredator tactics, we were surprised to notice a sex difference between male and female cane toads (Rhinella marina Linnaeus

1758) in the distribution of toxin-secreting glands across the body. First, casual observation suggests that glands are more numerous in the dorsal skin of adult males than females.

Second, field and laboratory observations indicate that stressed male toads sometimes exude toxin over most of their dorsal surface, whereas females mostly exude toxin from the parotoid glands. Lastly, when we attempted to extract toxin from the parotoids of toads (to use as an attractant in tadpole-trapping: Crossland et al., 2012), we obtained more exudate from the parotoids of females than of males. Those perplexing observations encouraged us to quantify the distribution of toxin-containing glands in male and female toads.

Histological studies have documented the morphology of bufonid skin glands in great detail (e.g. Hostetler & Cannon, 1974; Toledo & Jared, 1995; Schwinger et al., 2001; Jared et al., 2009). Most research has focused on the parotoid macroglands (which consist of around

120 to 130 secretory units: Hutchinson & Savitsky, 2004), to the neglect of smaller glands in other parts of the body. Nonetheless, the existence of these additional glands is well known.

For example, Regueira et al., (2016, p. 14–15) noted that “granular glands from the big warts in the skin of R. [Rhinella] arenarum produce toxins with similar characteristics to that of parotoid glands i.e., catecholamines and lipid-derived secretions, but do not display the same organisation as the macroglands”. The same authors indicated that “we did not observe differences between males and females for the studied skin regions [in the trunk region]”.

The ventral skin of bufonids contains mucous-producing but not toxin-producing glands

(Regueira et al., 2016), so was not included in our study.

180

Materials and Methods

Electro-Stimulation Trials

This component of the study was conducted on 13 adult toads, all from northeastern New

South Wales (Table 1). We conducted electro-stimulation trials on 16 standardised locations across the dorsal surface of each toad (to check that glands in the skin of the dorsum and limbs of both male and female toads exude toxin if stimulated, and measure the delay prior to secretion). Those locations were bilaterally symmetrical, with eight sites on each side of the body (dorsal surface of upper and lower forelimbs; dorsal surface of upper and lower hindlimbs; three evenly-spaced sites on either side of the dorsal midline; and parotoid macrogland).

We used a purpose-built electro-stimulator to deliver toxin (Lindley, 1969) using a maximum stimulation time of 240 s with 10 V and 130 Hz. During the trial period, we recorded the time prior to the first appearance of toxin (milky fluid) on the skin surface. If no toxin appeared within 240 s, we scored the response as “did not exude”.

Gland Attributes

For each toad used in the electro-stimulation trials, we also scored the following traits:

(1) Surface area and shape of the parotoid macroglands. – We photographed all toads and then used the image-analysis freeware ImageJ (Rasband, W.S., ImageJ, U.S. National

Institutes of Health, Bethesda, MD, USA) to calculate surface area of the macroglands. From the same photographs, we recorded maximum length and width (mm) of each parotoid macrogland.

(2) Proportion of the skin surface covered by glandular material. – We calculated the proportion of glandular material for each toad by pressing a transparent microscope slide marked with a 10  10 mm quadrat onto the toad’s body and scoring the proportion of that

181 quadrat that was glandular. These measurements were taken at the same locations where electro-stimulus trials were conducted. To confirm the presence of toxin-producing granular glands in the skin we also took histological samples from the dorsal tissues of a small subset of toads (four males and two females).

Field-Caught Animals

We also took the standardised measurements above on 238 adult cane toads (120 females,

118 males; 83 to 147 mm snout-urostyle length [SUL]), collected from 11 locations spanning the eastern to western edges of the species’ invasive distribution in Australia (Table 1). We included these geographically diverse samples to minimise impacts of any local variation, and to ensure that our results apply broadly. All patterns were consistent across populations, although the magnitude of sexual dimorphism sometimes differed. We do not describe these geographic effects in the present paper, because (together with sexual dimorphism in other traits) they are the subject of a separate study (Hudson et al. in prep.).

Captive-Raised Progeny

We repeated these morphological measurements on 134 adult (78 to 125 mm SUL) cane toads that had been raised in our field station from the egg stage (see Table 1 for details).

Field-caught adults had been spawned at the field station using hormonal priming, and the resultant offspring maintained under standard conditions (see Hudson et al. 2015 for more information on husbandry methods).

Statistical Analyses

Using JMP Pro version 11.2.0 (SAS Institute, Cary, NC), we conducted an ANOVA with toad sex and area of body (forelimb, hindlimb, dorsum, parotoid) as the factors and the delay

182 to exude toxin as the dependent variable. For analyses of morphological traits, we used an

ANCOVA with sex as the factor, toad body size (SUL) as the covariate, and the morphological measure (mean value per region per toad) as the dependent variable.

183

Table 1. Collection locations and specimens examined. This table shows data for wild-caught

adult toads (including specimens from Grafton, that were used for electro-stimulation trials as

well as morphological measurements); and also for captive-raised progeny of wild-caught

toads.

Female Male Mean Range Mean Range Site Latitude Longitude n (mm) (mm) n (mm) (mm) Wild-caught toads

Durack 15°56'43.74"S 127°13'16.81"E 9 110.8 92-128 10 105.2 87-117

Ellenbrae 15°58'27.64"S 127°03'43.70"E 12 116.8 106-134 5 116.2 110-120

Innisfail 17°31'28.85"S 146°01'56.38"E 10 113.3 105-120 12 110.3 100-121

Richmond 20°43'46.04"S 143°08'30.10"E 10 130.8 118-142 13 117.1 102-125

Katherine 14°26'20.39"S 132°16'18.67"E 14 130.3 115-144 11 121.4 102-132

Oombulgurri 15°10'49.50"S 127°50'42.14"E 14 119.2 111-136 11 112.0 107-117

Tully 17°55'58.33"S 145°55'24.80"E 11 112.9 94-147 10 98.9 83-114

Townsville 19°15'27.44"S 146°49'4.36"E 13 116.0 105-127 10 104.2 96-113

Mt. Isa 20°43'28.94"S 139°29'50.86"E 10 105.1 87-123 15 103.8 95-115

Jabiru 12°40'15.54"S 132°50'23.33"E 11 130.2 120-135 14 119.7 110-135

Brooms Head 29°36'29.65"S 153°20'08.83"E 6 109.0 100-123 7 103.5 93-120

Captive-raised progeny

El Questro 16°00'84.38"S 127°97'98.11"E 4 105.3 61-120 1 100 100

Innisfail 17°31'28.85"S 146°01'56.38"E 16 131.2 53-219 15 110 54-184

Oombulgurri 15°10'49.50"S 127°50'42.14"E 11 98.7 56-158 6 96.0 53-170

Purnululu 17°52'97.52"S 128°40'08.38"E 3 76.7 37-99 3 78.5 73-83

Townsville 19°15'27.44"S 146°49'4.36"E 13 116.9 58-203 16 96.4 52-138

Tully 17°55'58.33"S 145°55'24.80"E 14 123.1 69-194 23 94.2 50-175

Wyndham 15°46'48.03"S 128°10'01.43"E 3 90.9 61-129 6 104.8 75-152

184

Results

Electro-Stimulation Trials

All of the body parts we identified as “glandular” exuded milky fluid when stimulated, suggesting that these are all toxin-exuding glands (Figure 1).

The time taken from the onset of electro-stimulation to overt visual evidence of milky fluids being exuded ranged from 3 to 110 s (mean ± 1 SEM: 20.73 ± 1.75 s, n = 181 readings).

The time taken prior to exudation differed among areas of the body (forelimbs, hindlimbs, dorsum, parotoids) in different ways in males versus females (interaction sex*area of body,

F3,180 = 5.38, P < 0.002). That significant interaction term reflects a trend for dorsal glands to exude more rapidly in females than in males (Figure 2). If we analyse the data separately for each area of the body, the sex difference in mean time to exude toxin is significant for dorsal glands (females are quicker: F1,75 = 13.74, P < 0.001) and parotoids (males are quicker: F1,25

= 5.27, P < 0.04) but not for limbs (forelimbs, F1,34 = 2.94, P = 0.17; hindlimbs, F1,43 = 0.82,

P = 0.37).

Gland Attributes of Field-Caught Animals

Male and female toads differed in the relative size of the parotoid macroglands, and in the proportion of skin surface covered by toxin-secreting glands in other parts of the body. These patterns in sexual dimorphism were evident within all populations that we examined.

On average, males had smaller parotoid macroglands than did females both in absolute terms, and relative to body size (interaction sex*SUL, F1,210 = 4.52, P < 0.04; main effect of sex, F1,210 = 7.08, P < 0.01; see Figure 3a). Additionally, the parotoid glands of male toads were more elongate than were those of females (ANCOVA with sex as the factor,

185 macrogland width [mean of left and right glands] as the covariate, mean macrogland length as the dependent variable; interaction sex*SUL, F1,214 = 2.95, P = 0.09; sex effect, F1,214 =

10.98, P < 0.002; Figure 3b). However, the proportion of the body surface covered by glandular material was higher in males than females in other areas of the body. The sex disparity was significant for the dorsum (F1,215 = 76.57, P < 0.0001; Figure 4a), and hindlimbs (F1,215 = 28.84, P < 0.0001; Figure 4b) but not the forelimbs (F1,218 = 0.56, P = 0.45;

Figure 4c). Histological samples confirmed that both granular and mixed glands were present in the dorsal skin of both sexes (Figure 1).

Gland Attributes of Captive-Raised Animals

Our measurements of captive-raised progeny showed patterns almost identical to those in wild-caught animals (above). Specifically, males had smaller parotoid macroglands than did females (interaction sex*SUL, NS in all cases; main effect of sex, F1,130 = 8.33, P < 0.005), whereas the proportion of the body surface covered by glandular material was higher in males than females for the dorsum (F1,132 = 135.24, P < 0.0001), the hindlimbs (F1,132 = 24.78, P <

0.0001) and the forelimbs (F1,132 = 15.58, P < 0.0001).

As in field-caught animals, males had more elongate parotoids than did females

(ANCOVA with sex as the factor, macrogland width as the covariate, macrogland length as the dependent variable; interaction sex*SUL, F1,131 = 1.46, P = 0.23; sex effect, F1,131 = 5.05,

P < 0.03).

186

Figure 1. Histological sample of male dorsal skin depicting both granular (g) and mixed (m) glands and a male cane toad secreting toxin from its parotoid macroglands. Photographs by G.

Brown and C. Shilton (upper panel) and J. DeVore, M. Crossland, and C. Hudson (lower panel).

187

Figure 2. Delay prior to secretion of milky fluid from skin glands following standardised electro-stimulation of various parts of the body of male and female cane toads (Rhinella marina). The graph shows mean values and associated standard errors for toxin release times for glands on the toad’s dorsal surface and forelimbs and hindlimbs, as well as the parotoid macroglands.

188

Figure 3. The size (surface area) and shape of parotoid glands of male and female cane toads as a function of body length. Panels show (a) surface area of parotoid macroglands relative to toad body length, and (b) sex-based divergence in the shape (maximum length relative to maximum width) of parotoid macroglands in wild-caught cane toads, Rhinella marina. The graph shows mean values and associated standard errors for gland surface area for each 10- mm size class in snout-urostyle length. The extreme values (90 and 130 mm) include a few individuals slightly smaller (for 90 mm) or larger (for 130 mm) than the stated sizes.

189

Figure 4. Sex differences in the proportion of the cane toad skin covered in glandular material in sections of the (a) dorsum, (b) hindlimbs and (c) forelimbs.

190

Discussion

Despite an extensive (mostly, histologically-based) literature on the skin glands of bufonids, the existence of strong sexual dimorphism in the size and location of toxin-containing glands appears to have escaped notice. Our data show that male cane toads have smaller and more elongate parotoid macroglands than females of the same body size, but greater glandular coverage on the dorsum and limbs. All of these glands secrete milky fluid under electro- stimulation, suggesting that all are indeed reservoirs of toxin. Future work could usefully examine the chemical composition of the fluid released by different regions in the body in male and female toads; although it appeared similar (and was reported to be approximately the same chemically by Regueira et al., 2106), more detailed analysis would be of interest.

The amount of electrical stimulation required to elicit exudate differed among glands, and also differed (in some cases) between male and female toads stimulated at the same location on the body.

What mechanisms are responsible for these sex differences? Broadly, two non- exclusive processes might be at work: developmental plasticity and/or adaptation. Both are plausible a priori. First, experimental manipulations have revealed that many aspects of anuran morphology are highly plastic (Relyea, 2001; Kearney et al., 2014). Directly relevant to the current study, Hagman et al. (2009) reported that the size of the parotoid macroglands in metamorph cane toads is increased by exposure to alarm cues during larval life. Although male and female tadpoles presumably encounter similar environments, sensitivity to developmental conditions later in life might generate sex differences in gland sizes. Adult male and female cane toads utilise available habitats differently; for example, females are often found in densely vegetated sites far from water, whereas males congregate around breeding ponds (González-Bernal et al., 2015). Abiotic or biotic differences between their resultant experiences (e.g. in operative temperatures or in exposure to predation) might

191 directly modify morphological development, including investment into the parotoids and other glands. However, the magnitude and consistency of male-female differences, and the persistence of those differences even in captive-raised offspring (with no opportunity to select sex-specific habitats) argue strongly against this interpretation.

Sexual dimorphism in the distribution of toxin-containing glands thus appears likely to have evolved because of adaptation: that is, the sexes have evolved different spatial distributions of these glands across their bodies because of the way that gland size and distribution affects an individual toad’s viability. The factors generating that sex disparity in fitness consequences remain unclear, but we suggest the following possibilities:

(1) Sex differences in vulnerability to predation. – Adult male toads that call from exposed sites near water bodies may be vulnerable to avian predators (Beckmann & Shine,

2011), and hence benefit from a wide distribution of toxin across the dorsal surface (the part of the body most likely to be contacted by an aerial predator). In contrast, females that live in thicker vegetation may be more at risk from terrestrial predators such as rodents (González-

Bernal et al., 2015) and snakes (Shine, 2010), that are best repelled by a frontal display that exposes the parotoids and forelimb glands prominently.

(2) The need for females to redeploy toxin to developing eggs. – The eggs of cane toads contain high levels of several bufadienalides (Hayes et al., 2009). If these chemicals are synthesised in the parotoids from cholesterol precursors (Hutchinson & Savitsky, 2004), they may then need to be transferred through the bloodstream to the ovaries. Concentrating the site of toxin storage might facilitate such a redeployment.

(3) Male-male rivalry. – Male toads compete vigorously for access to females, with rival males often usurping an already-amplexed animal; those wrestling bouts likely confer strong selection on a male’s ability to cling to his partner (and thus explain the seasonal swelling of forearm musculature and nuptial spines in male anurans: Shine, 1979; Wells, 2010). The

192 dorsal skin of male R. marina becomes spinose during the breeding season, and may assist males to deter other male from displacing them during amplexus. Because the initially amplectant male clings firmly to the female, a newly-arriving male ends up on top of the first male – a position that causes the spinose skin of the first male to push directly against the thin ventral skin (including, pelvic patch) of the potential usurper. Exudates from glands in that area could further repel the rival; the relatively slow release of toxin from male dorsal glands fits well with this interpretation. Information on the chemical composition of these exudates

(irritants rather than toxins?) would be of great interest.

(4) Osmotic or hydric balance. – The contents of these glands include not only bufogenins, but also a diverse array of biogenic amines and glycosaminoglycans (Clarke,

1997). All of these substances may play wider biological roles in osmotic balance and/or hydric balance (Elkan, 1968; Vialli et al., 1969; Dapson, 1970; Lillywhite, 1971; Le Quang

Trong, 1975a,b; Matoltsy & Bereiter-Hahn, 1986; Lichtstein et al., 1992; Toledo et al., 1992;

Toledo & Jared, 1995). Thus the toxin-containing glands may serve additional functions other than deterrence of predators. In some anurans (but apparently not in cane toads: Schwinger et al., 2001), males and females differ in skin thickness (Greven et al., 1995, for Xenopus laevis); such a difference might impose selection on a range of functions that are conferred by glandular secretions.

A priori, we might expect an animal that deters predators by chemical means to deploy that defensive material widely across the body. Such a distribution would maximize the probability of a predator encountering the repellent regardless of where on the toad’s body it directs its initial attack. In that sense, the concentration of toxins within the parotoid macroglands of bufonids is surprising. However, at least some predators are likely to be detected by the toad before they launch an attack, allowing the toad to orient itself to face the oncoming predator, and inflate its lungs (both increasing its apparent size, and [by providing

193 firm pressure from beneath] facilitating emptying of toxin-storage glands under pressure from the predator: Mailho-Fontana et al., 2014). If toads can detect approaching predators in enough time for the anuran to orient towards the threat, a concentration of toxin in the anterior part of the body may maximise the arsenal exposed to the predator’s initial onslaught.

More broadly, interspecific as well as intraspecific (sex-based) divergences in the amount and location of toxin-storage glands warrant additional research. Within the “true toads” (Bufonidae), for example, morphologically similar species vary enormously in the relative size of the parotoid glands, and in whether or not toxin-secreting glands also occur on the limbs (Bücherl & Buckley, 1971; Blair, 1972). To our knowledge, that variation has never been examined in any comprehensive framework. Given the strong allometry of parotoid size in cane toads (Phillips & Shine, 2006), it would be of great interest to explore relationships between a species’ body size, ecology, and its investment into toxin production and storage.

Sex differences in size, ecology and toxin gland location equally would be worth investigating, as would intraspecific (geographic) variation in such traits. For example, do bufonids with smaller parotoids have more toxins distributed around their bodies, in smaller glands (as is the case with male versus female cane toads)?

For a comprehensive understanding of variation in investment into chemical defences, we also would need to quantify not only the store of toxin within a toad’s body, but also the rate at which that store can be replenished after it is used against a predator (see Jared et al.,

2014 for an example of this approach). Plausibly, a sex or species or life-history stage capable of replenishing toxin supplies more rapidly could “afford” to maintain a lower store of toxin. Future research also could explore factors that affect the toad’s willingness to exude its toxin stores under natural (or simulated) conditions. In our experience, cane toads generally rely on crypsis or escape if approached by a predator, and rarely express toxin unless they are under severe duress. However, individuals with severe spinal arthritis (and

194 hence, less capable of sustained locomotion) soon cease attempting to move away from the threat and instead resort to defensive displays with toxin secretion (Brown et al., 2007). It would be relatively straightforward to investigate the effects of a toad’s sex, size, location, and previous experience, as well as weather conditions, in determining the animal’s propensity to deploy toxins at the skin surface when harassed. In short, we know a great deal about the detailed morphology of toxin-storage glands in anurans, but far less about the ways in which the animals actually use those toxins for defence against predators. The strong sexual dimorphism in location of toxin stores in cane toads is intriguing, and hints that ecological factors may influence the tactics used by anurans to store and deploy their formidable chemical weaponry.

195

Acknowledgements

We thank Greg Brown, Matthew Greenlees and Michael Crosssland for collecting samples, and Melanie Elphick and Chalene Bezzina for laboratory assistance. The work was funded by the Australian Research Council (FL120100074), the National Natural Science Foundation of

China (31670392) and the China Scholarship Foundation (201408515128).

Literature Cited

Andersson, M.B. 1994 Sexual Selection. Princeton University Press, Princeton, NJ.

Beckmann, C., Shine, R. 2011. Toad's tongue for breakfast: exploitation of a novel prey type,

the invasive cane toad, by scavenging raptors in tropical Australia. Biol. Invasions 13:

1447–1455.

Blair, W.F. 1972. Evolution in the Genus Bufo. University of Texas Press, Austin, TX.

Brodie Jr., E.D., Ridenhour, B.J., Brodie III, E.D. 2002. The evolutionary response of

predators to dangerous prey: hotspots and coldspots in the geographic mosaic of

coevolution between garter snakes and newts. Evolution 56: 2067–2082.

Brown, G.P., Shilton, C.M., Phillips, B.L., Shine, R. 2007. Invasion, stress, and spinal

arthritis in cane toads. Proc. Natl. Acad. Sci. USA 104: 17698–17700.

Bücherl, W., Buckley, E.E. 1971. Venomous Animals and Their Venoms. Volume II.

Venomous Vertebrates. Academic Press, New York.

Carazo, P., Noble, D.W.A., Chandrasoma, D., Whiting, M.J. 2014. Sex and boldness explain

individual differences in spatial learning in a lizard. Proc. Roy. Soc. B 281: 20133275.

Clarke, B.T. 1997. The natural history of amphibian skin secretions, their normal functioning

and potential medical applications. Biol. Rev. Camb. Philos. Soc. 72: 365–379.

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

Princeton, NJ.

196

Crossland, M.R., Haramura, T., Salim, A.A., Capon, R.J., Shine, R. 2012. Exploiting

intraspecific competitive mechanisms to control invasive cane toads (Rhinella marina).

Proc. Roy. Soc. B 279: 3436–3442.

Dapson, R.W. 1970. Histochemistry of mucus in the skin of the frog Rana pipiens. Anat. Rec.

166: 615–625.

Doucet, S.M., Mennill, D.J. 2009. Dynamic sexual dichromatism in an explosively breeding

Neotropical toad. Biol. Lett. 6: 63–66.

Elkan, E. 1968. Mucopolysaccharides in the anuran defense against desiccation. J. Zool. 155:

19–53.

Emerson, S.B. 2001. A macroevolutionary study of historical contingency in the fanged frogs

of Southeast Asia. Biol. J. Linn. Soc. 73: 139–151.

Fairbairn, D.J., Blanckenhorn, W.U., Székely, T. 2007. Sex, Size and Gender Roles:

Evolutionary Studies of Sexual Size Dimorphism. Oxford University Press, New York.

González-Bernal, E., Brown, G.P., Crowther, M.S., Shine, R. 2015. Sex and age differences

in habitat use by invasive cane toads (Rhinella marina) and a native anuran (Cyclorana

australis) in the Australian wet-dry tropics. Austral Ecol. 40: 953–961.

Greven, H., Zanger, K., Schwinger, G. 1995. Mechanical properties of the skin of Xenopus

laevis (Anura, Amphibia). J. Morphol. 224: 15–22.

Hagman, M., Hayes, R.A., Capon, R.J., Shine, R. 2009. Alarm cues experienced by cane toad

tadpoles affect post-metamorphic morphology and chemical defences. Funct. Ecol. 23:

126–132.

Hayes, R.A., Crossland, M.R., Hagman, M., Capon, R.J., Shine, R. 2009. Ontogenetic

variation in the chemical defenses of cane toads (Bufo marinus): toxin profiles and

effects on predators. J. Chem. Ecol. 35: 391–399.

197

Hostetler, J.R., Cannon, M.S. 1974. The anatomy of the parotoid gland in Bufonidae with

some histochemical findings. I. Bufo marinus. J. Morphol. 142: 225–239.

Hudson, C.M., He, X., Fu, J. 2011 Keratinized nuptial spines are used for male combat in the

Emei Moustache Toad (Leptobrachium boringii). Asian. Herpetol. Res. 2: 142–148.

Hudson, C.M., Phillips, B.L., Brown, G.P., Shine, R. 2015. Virgins in the vanguard: low

reproductive frequency in invasion-front cane toads. Biol. J. Linn. Soc. 116: 743–747.

Hutchinson, D.A., Savitzky, A.H. 2004. Vasculature of the parotoid glands of four species of

toads (Bufonidae: Bufo). J. Morphol. 260: 247–254.

Jared, C., Antoniazzi, M.M., Jordao, A.E., Silva, J.R.M., Greven, H., Rodrigues, M.T. 2009.

Parotoid macroglands in toad (Rhinella jimi): their structure and functioning in passive

defence. Toxicon 54: 197–207.

Jared, S.G.S., Jared, C., Egami, M.I., Mailho-Fontana, P.L., Rodrigues, M.T., Antoniazzi,

M.M. 2014. Functional assessment of toad parotoid macroglands: a study based on

poison replacement after mechanical compression. Toxicon 87: 92–103.

Katsikaros, K., Shine, R. 1997. Sexual dimorphism in the tusked frog, Adelotus brevis (Anura:

Myobatrachidae): the roles of natural and sexual selection. Biol. J. Linn. Soc. 60: 39–51.

Kearney, B.D., Pell, R.J., Byrne, P.G., Reina, R.D. 2014. Anuran larval developmental

plasticity and survival in response to variable salinity of ecologically relevant timing

and magnitude. J. Exp. Zool. A 321: 541–549.

Kupfer, A. 2007. Sexual size dimorphism in amphibians: an overview. Pages 50–59 in D.J.

Fairbairn, W.U. Blanckenhorn, T. Székely, eds., Sex, Size and Gender Roles:

Evolutionary Studies of Sexual Size Dimorphism, Oxford University Press, Oxford, UK.

Le Quang Trong, Y. 1975a. La peau et les glandes cutanées de Dicroglossus occipitalis

Günther. Ann. Univ. Abidjan. Ser. E 8: 15–29.

198

Le Quang Trong, Y. 1975b. E’tude de la peau et des glandes cutanées de quelques

amphibiens du genre Ptychadena. Ann. Univ. Abidjan. Ser. E 8: 31–52.

Lee, J.C. 2001. Evolution of a secondary sexual dimorphism in the toad, Bufo marinus.

Copeia. 2001: 928–935.

Lichtstein, D., Gati, I., Haver, E., Katz, U. 1992. Digitalis-like compounds in the toad Bufo

viridis: tissue and plasma levels and significance in osmotic stress. Life. Sci. 51: 119–

128.

Lillywhite, H.B. 1971. Thermal modulation of cutaneous mucus discharge as a determinant

of evaporative water loss in the frog Rana catesbeiana. Z. Vgl. Physiol. 73: 84–104.

Lindley, B.D. 1969. Nerve stimulation and electrical properties of frog skin. J. Gen. Physiol.

53: 427–449.

Mailho‐ Fontana, P.L., Antoniazzi, M.M., Toledo, L.F., Verdade, V.K., Sciani, J.M., Barbaro,

K.C., Pimenta, D.C., Rodrigues, M.T., Jared, C. 2014. Passive and active defense in

toads: the parotoid macroglands in Rhinella marina and Rhaebo guttatus. J. Exp. Zool.

A 321: 65–77.

Matoltsy, A.G., Bereiter-Hahn, J. 1986. Introduction. Pages 1–7 in J. Bereiter-Hahn, A.G.

Matoltsy, K.S. Richards, eds., Biology of the Integument, Volume 2. Vertebrates.

Springer, New York.

Navas, C.A., James, R.S. 2007. Sexual dimorphism of extensor carpi radialis muscle size,

isometric force, relaxation rate and stamina during the breeding season of the frog Rana

temporaria Linnaeus 1758. J. Exp. Biol. 210: 715–721.

Neuhaus, P., Ruckstuhl, K.E. 2005. Sexual Segregation in Vertebrates. Cambridge University

Press, Cambridge, UK.

Phillips, B.L., Shine, R. 2006. Allometry and selection in a novel predator-prey system:

Australian snakes and the invading cane toad. Oikos 112: 122–130.

199

Regueira, E., Dávila, C., Hermida, G.N. 2016. Morphological changes in skin glands during

development in Rhinella arenarum (Anura: Bufonidae). Anat. Rec. 299: 141–156.

Relyea, R.A. 2001. Morphological and behavioral plasticity of larval anurans in response to

different predators. Ecology 82: 523–540.

Schwinger, G., Zanger, K., Greven, H. 2001. Structural and mechanical aspects of the skin of

Bufo marinus (Anura, Amphibia). Tissue Cell 33: 541–547.

Shine, R. 1979. Sexual selection and sexual dimorphism in the Amphibia. Copeia 1979: 297–

306.

Shine, R. 1989. Ecological causes for the evolution of sexual dimorphism: a review of the

evidence. Q. Rev. Biol. 64: 419–464.

Shine, R. 2010. The ecological impact of invasive cane toads (Bufo marinus) in Australia. Q.

Rev. Biol. 85: 253–291.

Toledo, R.C., Jared, C. 1995. Cutaneous granular glands and amphibian venoms. Comp.

Biochem. Physiol. 111: 1–29.

Toledo, R.C., Jared, C., Brunner, A. 1992. Morphology of the large granular alveoli of the

parotoid glands in toad (Bufo ictericus) before and after compression. Toxicon 30: 745–

753.

Tsuji, H., Matsui, M. 2002. Male-male combat and head morphology in a fanged frog (Rana

kuhlii) from Taiwan. J. Herpetol. 36: 520–526.

Vialli, M., Bolognani, L., Croce, G., Bolognani, F.A.M. 1969. Ricerche istochimiche e

biochimiche sugli esosaminglicani (MPS) del “veleno” delle parotoidi di Bufo vulgaris

e viridis. Arch. Biol. Sci. 53: 108–125.

Vincent, S.E., Shine, R., Brown, G.P. 2005. Does foraging mode influence sensory

modalities for prey detection? A comparison between male and female filesnakes

(Acrochordus arafurae, Acrochordidae). Anim. Behav. 70: 715–721.

200

Wells, K.D. 2010. The Ecology and Behavior of Amphibians. University of Chicago Press,

Chicago, IL.

201