Ecological mechanisms and evolutionary patterns of introduced

Habacuc Flores Moreno

Thesis submitted for the degree of Doctor of Philosophy Evolution & Ecology Research Centre School of Biological, Earth & Environmental Sciences University of New South Wales August 2013 PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Flores Moreno

First name: Habacuc Other name/s:

Abbreviation for degree as given in the University calendar: PhD

School: School of Biological, Earth and Environmental Sciences Faculty: Faculty of Science

Title: Ecological mechanism and evolutionary patterns of introduced species

Abstract 350 words maximum: (PLEASE TYPE) It has long been assumed that introduced species have higher seed dispersal and survival than do native species. These assumptions are central to our understanding of introduced species’ fast spread rates, competitive ability, and dominance in their new environments. However, studies show mixed results when comparing proxy traits for dispersal and survival. In this thesis, using data collected from the global literature, I compared seed dispersal and recruitment success (survival through germination, one week survival after germination, and survival from germination to first reproduction) of introduced and native species. Against all expectations, I found no significant difference between introduced and native species’ seed dispersal distance or recruitment success. I then explored the relative importance of nature (species’ level of invasiveness) and nurture (introduced vs. native ranges) on invasive species’ life history characteristics. I found that invasive species have higher seed production due to both their nurture and nature, but no differences in survival or dispersal. Lead by evolutionary theory, we assume that introduced species’ fast phenotypic changes could lead to diversification events. Nevertheless, we know little about the trajectories of changes in these species. I used herbarium specimens to track changes in area, leaf shape and height for three annual introduced species. I found that introduced species keep changing even 200 years after introduction and that across species there is a lag-phase in the rates of phenotypic change. Although introduced species sometimes show incredible dominance over native species, there is no fundamental reason to expect native species to be at a general disadvantage in dispersal or survival. It is possible that introduced species’ advantages such as higher seed production, enemy release or increase competitive ability might be countered, or at least matched, by native species superior adaptation to local conditions. However, that our three introduced species demonstrate rapid changes in phenotype after a lag phase suggests that some invaders may yet demonstrate their full potential. Overall, this thesis reshapes our knowledge about introduced species’ ecological strategies and shows that trait-based studies are deceiving when used as a proxy for processes, such as dispersal, recruitment and survival.

Declaration relating to disposition of project thesis/dissertation

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

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…………………………………………………………… ……………………………………..……………… ……….……………………...…….… Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS

ORIGINALITY STATEMENT

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

Signed ……………………………………………......

Date ……………………………………………......

Statement of contribution of co-authors and declarations of

permission to publish.

Chapters one to four comprise for standalone paper that have been prepared for publication in peer-reviewed scientific journals. Each chapter is self contained including introduction, methods, results, tables, figures, discussion, acknowledgments, references and appendices. The formatting of each chapter is in the style of the journal to which it has been/will be submitted. The contributions by co-authors for each chapter are listed below.

Chapter one-

Flores-Moreno, H., Thomson, F.J., Warton, D.I. and Moles, A.T. Are introduced species better dispersers than native species? A global comparative study of seed dispersal distance. PLoS ONE 8(6): e68541. doi:10.1371/journal.pone.0068541

The study was conceived and designed by HFM and ATM, HFM and FJT performed the experiments and collated the data. HFM, DIW and ATM analysed the data. DIW provided analytical tools. The manuscript writing was led by HFM, but all co-authors contributed to it.

i

Chapter two-

Flores-Moreno, H., and Moles, A.T. A comparison of the recruitment success of introduced and native species under natural conditions. PLoS ONE 8(8): e72509. doi:10.1371/journal.pone.0072509

The study was conceived and designed by HFM and ATM. HFM performed the experiments and collated the data. HFM and ATM analysed the data. The manuscript writing was led by HFM, but ATM contributed substantially to it.

Chapter three-

Flores-Moreno, H., Thomson, F.J., Dalrymple, D.L. and Moles, A.T. Nature vs.

Nurture: Are introduced species successful because of what they are, or where they are?

The study was conceived and designed by HFM, RLD and ATM. HFM performed the experiments.HFM and FJT collated the data. HFM, FJT and ATM provided data. HFM,

RLD and ATM analysed the data. The manuscript writing was led by HFM, but all co- authors contributed to it.

ii

Chapter four-

Flores-Moreno, H., Garcia Treviño, E., Letten, A.D., and Moles, A.T In the beginning:

Phenotypic change of three introduced species through their first 200 years since introduction.

The study was conceived and designed by HFM and ATM. HFM, EGT and ADL performed the experiments.HFM collected and collated the data. HFM, ADL and EGT analysed the data. EGT provided analytical tools. The manuscript writing was led by

HFM, but ATM and EGT contributed substantially to it.

iii

Acknowledgements

It is extremely difficult to express, in so few lines, my infinite gratitude to all the people that somehow helped me to get in, to get by and to finalize this project.

First of all, I would like to deeply thank my supervisor, Angela Moles. Without her, this thesis wouldn’t have been possible. Angela, in the last few years you have taught me a lot. Thank you for teaching me to write. Thank you for teaching me about the perks and risks of science. Thank you for helping me to articulate so many half cook ideas. Thank you for letting me fail and teaching me how to get through it. Most importantly, thank you for teaching me to do better science and to be a better scientist! I will be always grateful to you for your generosity both as a person and as a scientist.

Rhiannon, this thesis is as much yours as it is mine (you should let the GRS know, maybe they’ll give us two for one!). Thank you for being my editor, reviewer, co- author, psychologist, research assistance, friend and partner. This ride wouldn’t have been half as fun and probably wouldn’t have been possible without you. Thank you for the good times. Thank you for sticking around in the bad ones. Thank you for pitching ideas with me very early in the morning or very late at night. Thank you for proof- reading my drafts hundreds of times. Thank you for helping me learn to be a better person. Most importantly, thank you for always being there! I hope that never changes.

iv

A mi familia: mamá (Eliud “grande”), papá (Jose) y Eliud “chica”. Gracias por su apoyo incondicional. Mamá gracias por siempre creer en mí, siempre apoyarme y siempre sacarme de apuros. No sé cómo podría haber sobrevivido la adolescencia sin tu orientación. Papá, gracias por ayudarme con todas las estadísticas y problemas de índole numérico, pero sobretodo muchas gracias por enseñarme a siempre exigirme más, a ser paciente, a ser un buen hombre y sobre todo a ser un hombre razonable (excepto atrás del volante). Papá, gracias por enseñarme a ser humilde, eres un gran hombre. Eliud gracias por tu compañía, eres la mejor hermana que tengo. Gracias por siempre tener tiempo para mí, gracias por enseñarme a ser –o por lo menos conducirme como – una persona razonable. Eliud, gracias por enseñarme a ser considerado y a ser reflexivo para conmigo y para con los demás. Es difícil estar lejos de ustedes, pero eso me ha permitido ver cuánto han influenciado mi forma de ser y de actuar. A los tres gracias por hacerme un buen ciudadano de este mundo!

Jonci, I don’t know where to start... You are a wonderful friend. Thank you for sharing so many experiences with me. Thank you for all the chats and accompanying beers.

Thank you for all the delicious dinners! Thank you for helping me put things into perspective and to teach me that it is always important to find a way and time to have fun, and that there is always room for improvement.

v

To Melanie and Chris thank you for taking me into your family, I hope that you don’t regret it. Thank you for all the breakfast, dinners, lunches and . Thank you for all your support and the warm affection.

To the Big Ecology lab members, present and past - Fiona, Ray, Laura, Rhiannon,

Ellen, Marianne, Sichong, Claire, Flo, Tom and Tim. Thank you for the chats, coffees and help. Thank you for your support you are an awesome research group.

Margo thank you for always organizing fun stuff and teaching me that things are not always as good as they look, or as bad –fo r that matter. Thank you for all the chats about everything and nothing. Thank you for all the dinners and all the fun times. It was fun to discover High tea with you.

Rowena, Nico, Marie, Christian, Mike, Jene, Jacob, Kaz, Bridge and Mark Brown - thank you for all the bbq’s, climbing trips, fishing trips, birthday trips, dinners and pub nights. It is awesome to have you around, it is awesome to be friends with you. Thank you for keeping me fit and well fed. Thank you for reminding me that there is more to life than science and keeping me earth grounded.

Andrew, thank you for always arguing with me about science, it has been lots of fun.

But also thank you for your friendship and all those coffee chats that lead to nowhere in particular.

vi

Eddie, Sam, Jo O’Cock, Alyssa, Heather, Sylvia and all the BEES people than you for your friendship and making life enjoyable in campus. Sam and Jo O’Cock thank you for the morning teas and good times.

vii

Abstract

It has long been assumed that introduced species have higher seed dispersal and survival than do native species. These assumptions are central to our understanding of introduced species’ fast spread rates, competitive ability, and dominance in their new environments. However, studies show mixed results when comparing proxy traits for dispersal and survival. In this thesis, using data collected from the global literature, I compared seed dispersal and recruitment success (survival through germination, one week survival after germination, and survival from germination to first reproduction) of introduced and native species. Against all expectations, I found no significant difference between introduced and native species’ seed dispersal distance or recruitment success. I then explored the relative importance of nature (species’ level of invasiveness) and nurture (introduced vs. native ranges) on invasive species’ life history characteristics. I found that invasive species have higher seed production due to both their nurture and nature, but no differences in survival or dispersal. Lead by evolutionary theory, we assume that introduced species’ fast phenotypic changes could lead to diversification events. Nevertheless, we know little about the trajectories of changes in these species. I used herbarium specimens to track changes in leaf area, leaf shape and plant height for three annual introduced species. I found that introduced species keep changing even 200 years after introduction and that across species there is a lag-phase in the rates of phenotypic change. Although introduced species sometimes show incredible dominance

viii

over native species, there is no fundamental reason to expect native species to be at a general disadvantage in dispersal or survival. It is possible that introduced species’ advantages such as higher seed production, enemy release or increase competitive ability might be countered, or at least matched, by native species superior adaptation to local conditions. However, that our three introduced species demonstrate rapid changes in phenotype after a lag phase suggests that some invaders may yet demonstrate their full potential. Overall, this thesis reshapes our knowledge about introduced species’ ecological strategies and shows that trait-based studies are deceiving when used as a proxy for processes, such as dispersal, recruitment and survival.

ix

Table of contents

Statement of contribution i

Acknowledgements iv

Abstract viii

Introduction 1

Chapter one- Are introduced species better dispersers than native species? A global comparative study of seed dispersal distance. 19

Chapter two- A comparison of the recruitment success of introduced and native species under natural conditions 65

Chapter three- Nature vs. Nurture: Are introduced species successful because of what they are, or where they are? 115

Chapter four- In the beginning: Phenotypic change of three introduced species through their first 200 years since introduction. 153

Conclusions 192

x

Introduction

I come from a population ecology background. Population demographic models are of much interest to me, as with them we are able to trace the evolution of different life histories strategies and measure (to some extent) the fitness of populations and their individuals. My interest in fitness and the evolution of life history strategies brought me to study introduced species. This is because with some horror, but mostly intrigue, I have long observed introduced species and wondered how they seem to maximise almost all aspects of fitness and out-compete native species. It seems impossible, but some introduced species seem to defy natural laws and simply take over ecosystems around the world. Examples of these are the Nile perch in Tanzania, the cane toad in

Australia, and the kudzu in North America. When I started this thesis it had already been proposed that introduced species were better dispersers, had higher recruitment, grew faster, had higher reproductive output and in general achieved higher performance than did native species. I thought that, standing on years of previous research, my thesis would contribute to the field by quantifying the advantage of introduced species compared to native species.

There are a wide range of prospective hypotheses to explain introduced species’ success

(Elton, 1958; Pyšek et al., 1994; Blossey and Notzold, 1995; Lonsdale, 1999;

Simberloff and von Holle, 1999; Davis et al., 2000; Keane and Crawley, 2002;

Callaway and Ridenour, 2004; Colautti et al., 2004; Levine et al., 2004; Hierro et al.,

1 Introduction

2005; Alpert, 2006; Colautti et al., 2006; Richardson and Pyšek, 2006; Sax et al., 2007; summarized in Catford et al., 2009). But, to some degree, all of the hypotheses can be divided into two groups: those that focus on the role of the novel environment and those that focus on the role of species’ traits.

Introduced species and the role of traits

Baker (1974) suggested a list of traits and life history characteristics that were associated with “weedy” species. This list of traits was quickly adopted by invasion ecology probably because, similar to weedy species, introduced species tend to spread through the landscape, dominate communities and have high population growth rates.

Current theory suggests that introduced species tend to have short generation times, high fecundity, high growth rates, higher seed production, better dispersal abilities and high recruitment success (summarized in Lodge, 1993; Sakai et al., 2001; Pyšek and

Richardson, 2007). Most of the traits underlying introduced species’ success are correlated with these life history characteristics.

Traits that have been commonly compared between introduced and native species are growth form, life form, specific leaf area, flowering phenology, dispersal mode, plasticity, clonality, rapid germination and seedling emergence (summarized in Pyšek and Richardson, 2007). Plant height and seed mass are a crucial part of species’ life history strategies (Westoby et al., 1992; Aarssen and Jordan, 2001; Henery and

Westoby, 2001; Westoby et al., 2002; Moles et al., 2009), and are among the most studied traits across introduced species. However, studies comparing seed mass and plant height between introduced and native species have arrived at contradictory results.

2 Introduction

Some studies have shown that introduced species are smaller and/or have smaller seeds

(Vilà et al., 2005; Ordonez et al., 2010). However several studies have reported greater seed mass or taller in introduced species (Daws et al., 2007; Hawkes, 2007), no difference between introduced and native plants in height and seed mass (Thompson et al., 1995; Mason et al., 2008) or even mixed patterns across regions, populations or between and within species (Thébaud and Simberloff, 2001; Buckley et al., 2003;

Maron et al., 2004). Previous research has also returned contrasting results when comparing other traits between introduced and native species. For instance, Davidson et al. (2011) found that on average introduced species have higher plasticity than native species, but other studies have found no difference in plasticity between introduced and native species (Godoy et al., 2011; Palacio-López and Gianoli, 2011). Similarly, other studies have found contradictory results when comparing traits like dispersal syndrome

(Thompson et al., 1995; Cadotte and Lovett-Doust, 2001; Prinzing et al., 2002; Lake and Leishman, 2004), growth rate (Bellingham et al., 2004; Grotkopp and Rejmánek,

2007; Dawson et al., 2011) or rapid germination and seedling emergence (Garcia-

Serrano et al., 2004; van Kleunen and Johnson, 2007; Schlaepfer et al., 2010; Chrobock et al., 2011) between introduced and native species. Overall, there is well grounded evidence that suggests that invasive-introduced and native species have differences in performance (van Kleunen, Weber, et al., 2010), but there seems to be a lack of consensus on the direction and intensity of the differences between introduced and native species.

The contradictory results stemming from the study of introduced species’ traits have, for a long time, generated debate in the field about their importance in characterizing invasive species (Williamson, 1999; Thompson and Davis, 2011; van Kleunen et al.,

3 Introduction

2011). However, we do often rely on the assumption that introduced species’ traits positively affect their fitness. As Ackerly & Monson (2003) eloquently put it (referring to the study of plant function) “we often accept the adaptive nature of morphological and functional traits despite little or no evident connection between function and fitness”. This situation of course does not only apply to invasion ecology, but it is pervasive in this discipline. Rather than adding another case study to the mass of data already available, I aimed to do something new: to bypass the use of traits as proxies to infer fitness and actually quantify differences in performance between introduced and native species. By comparing the differences in life history characteristics between introduced and native species I hoped to be able to characterize where in the life cycle introduced species outcompete native species.

Introduced species and the role of range

Elton (1958) was the first to establish that the characteristics of a community would determine its’ susceptibility to introduced species. Since Elton’s contribution, the field has moved forward quantifying the community effects in space (e.g. interactions, change in resource availability, empty niches), and time (niche shift, naturalization, species’ rapid evolution) in introduced species success. Many current theories in invasion ecology are based on the idea that introduced species will increase their performance in their introduced range as a result of the positive effect of the new dynamics and interactions that they experience. For instance, it has been proposed that introduced species could benefit from facilitative interactions with other invaders

(Simberloff and von Holle, 1999; Simberloff, 2006), or from enemy release in their new range (Keane and Crawley, 2002; Liu and Stiling, 2006; but see Chun et al., 2010).

4 Introduction

Introduced species could go through evolution of increased competitive ability as a result of forming new positive interactions in their introduced range or leaving negative ones behind (Blossey and Notzold, 1995; Bossdorf et al., 2005; but see Felker-Quinn et al., 2013). Recently, a study by Parker et al. (2013) compared introduced and native populations across 57 species finding that not all introduced species are expected to perform better in their introduced range ( Parker et al., 2013). However, on average, individuals from introduced populations tend to be larger and more fecund than their native range counterparts.

From the available evidence (Grigulis et al., 2001; Buckley et al., 2003; Maron et al.,

2004; Jongejans et al., 2008; Hierro et al., 2013), we would expect species to have greater seed dispersal distance, recruitment success and seed production in their introduced range than in their native range. Unfortunately, as the majority of the studies dealing with introduced species ranges have been focused on size, fecundity and abundance related traits (summarized in Parker et al., 2013), we know little about the effect of range on other crucial life history characteristics and traits of introduced species. Differences between species growing in their introduced and native range could reflect differences that evolved or were enhanced by the range shift, or could reflect differences due to intrinsic characteristics of the species under comparison (van

Kleunen, Dawson, et al., 2010). Most previous research has been unable to disentangle the effect of range and the effect of species’ intrinsic characteristics. This is because most studies focus on the effects of range (Blossey and Notzold, 1995; Buckley et al.,

2003; Hierro et al., 2005; Jongejans et al., 2008) or species characteristics (Rejmánek,

1996; Williamson and Fitter, 1996; Burns et al., 2013), but not both. In this thesis I hoped to be able to disentangle the effect of range shift and introduced species’ intrinsic

5 Introduction

characteristics on seed production, seed dispersal, survival through germination and early seedling survival.

Standing on the shoulders of giants: data synthesis

There are two main quantitative approaches to conducting large scale comparative analyses across studies; meta-analysis and data compilations. Both methods summarize the results from two or more studies however they differ in the way they do it. In the last two decades meta-analyses have become a common tool in ecology and evolution

(Nakagawa and Poulin, 2012; Vetter et al., 2013), because they allow higher statistical power (Higgins et al., 2008). However, meta-analyses are affected by the correct calculation of a effect size index or response ratio, the correct weighting of the effect sizes, and the pooling of the effect sizes into different summary effects (Hedges et al.,

1999; Nakagawa and Poulin, 2012; Vetter et al., 2013). These requirements can preclude many studies with heterogeneous approaches, considerably reducing the sample size. Data compilations is a method in which primary data from different studies is collated into a single database. Data compilations have been a common tool in comparative ecology and evolution (e.g. Silvertown et al., 1993; Moles and Westoby,

2006; Kattge et al., 2011) because analysis of primary data can be performed across studies providing crucial information on common biological patterns across species

(Pugnaire and Valladares, 2007). Data compilations provide vital information to understand broad scale patterns or general laws across species or groups of species.

I decided to use a data compilation approach because the purpose of this thesis was to look for general rules in introduced species ecological strategies and be able to

6 Introduction

quantitatively compare the variation in introduced and native species characteristics.

Thousands of studies on a single or a few introduced species’ characteristics have accumulated through the years. These observations are of great value, and generalizing across them allows us to have a better picture of the ecological variation of introduced and native species. Furthermore, comparisons made across a reduced number of species or within a specific region would provide relevant information for the targeted species or regions. However extrapolating the information from a local or regional study when searching for general rules in the ecological strategies of introduced species would require a certain level of inference. Sixty to 98% of trait variance in plants occurs at the between species level (Kattge et al., 2011). Because of this, and the broad scale focus of my thesis, I decided to assess the difference between introduced and native organisms at the cross species level. However, only primary studies including more than ten replicates per species have been included. This is because ten replicates per species is often deemed as an appropriate level of replication for comparing life history characteristics across species without sacrificing accuracy for efficiency (Cornelissen et al., 2003)

The main aim of this thesis is to quantify the difference in life history characteristics of introduced and native species throughout their life cycle on a broad scale and always under natural conditions. I decided to do these comparisons under natural conditions because although comparisons under experimental conditions show us whether introduced and native species differ, everything else being equal, they do not tell us how introduced species are dealing with natural selective pressures and interactions with the surrounding environment.

7 Introduction

Terminology

Across studies tends to be a huge variation in the terminology used by individual researchers (Richardson et al., 2000; Colautti and Richardson, 2009). Furthermore, studies comparing introduced and native species answer different questions that those comparing species by their level of invasiveness (van Kleunen, Dawson, et al., 2010).

Variation in terminology and variation between comparative approaches could generate spurious or contradictory results in the comparison of introduced and native species or of species by their level of invasiveness. To avoid any potential problems with terminology throughout this thesis I use two categories for species’ status: introduced species are species that are known to be introduced anywhere in the world and native species are species that are distributed in their native range. I also use four categories of invasiveness: Invasive species have rapid population increase and rapid range expansion

(sensu Richardson et al., 2000). Naturalized species establish self-replacing populations without range expansion, while casual aliens do not form permanent self-replacing populations in their area of introduction (sensu Richardson et al., 2000).Untransported species are native species uniquely distributed in their region of origin, not known to be introduced or invasive anywhere in the world.

Thesis structure

Each of the chapters of this thesis attempts to describe the ecological mechanisms underpinning the success of introduced species on a broad spatial or temporal scale and always under natural conditions.

8 Introduction

Chapter 1 tests whether introduced species have longer seed dispersal distance (mean and maximum) than native species at a global scale under natural conditions.

Chapter 2 tests whether introduced species have higher survival through germination, early seedling survival and survival from germination to first reproduction than native species at a broad scale under natural conditions.

Chapter 3 compares the relative effect of species’ level of invasiveness and range shift

(introduced vs. native ranges) on seed production, seed dispersal (mean and maximum) survival through germination and early seedling survival cross species on a broad scale under natural conditions.

Chapter 4 tests whether introduced species phenotypic change goes through a lag phase shortly after introduction and whether introduced species still express phenotypic change hundreds of years after their introduction.

All data chapters for this thesis have been written for publication in scientific journals.

The first two chapters have already been published in PLoS One. Because each chapter is written as a standalone piece, there is some inescapable degree of overlap between chapters. The first person plural “we” and the possessive pronounce “ours” are used throughout the data chapters as none of them are sole-authored. However, I am mostly

9 Introduction

guilty for the concepts, the majority of the data collection, data analysis and writing involved in each of the manuscripts (see statement of contributions, page i for details).

The formatting of each chapter is in the style of the journal to which it has been/will be submitted. Co-authors are listed in each data chapter, and so are acknowledgements to person/institutions that in some way contributed to the realization of each chapter.

References

Aarssen, L. W. & Jordan, C. Y. 2001. Between-species patterns of covariation in plant

size, seed size and fecundity in monocarpic herbs. Ecoscience, 8: 471-477.

Ackerly, D. D. & Monson, R. K. 2003. Waking the Sleeping Giant : the evolutionary

foundations of plant function. International Journal of Plant Sciences, 164: S1-

S6.

Alpert, P. 2006. The advantages and disadvantages of being introduced. Biological

Invasions, 8: 1523-1534.

Baker, H. G. 1974. The evolution of weeds. Annual Review of Ecology and Systematics,

5: 1-24.

Bellingham, P. J., Duncan, R. P., Lee, W. G. & Buxton, R. P. 2004. Seedling growth

rate and survival do not predict invasiveness in naturalized woody plants in New

Zealand. Oikos, 106: 308-316.

Blossey, B. & Notzold, R. 1995. Evolution of increased competitive ability in invasive

nonindigenous plants: A hypothesis. Journal of Ecology, 83: 887-887.

Bossdorf, O., Auge, H., Lafuma, L., Rogers, W. E., Siemann, E. & Prati, D. 2005.

Phenotypic and genetic differentiation between native and introduced plant

populations. Oecologia, 144: 1-11.

10 Introduction

Buckley, Y. M., Downey, P., Fowler, S. V., Hill, R., Memmot, J., Norambuena, H.,

Pitcairn, M., Shaw, R., Sheppard, A. W., Winks, C., et al. 2003. Are invasive

bigger? A global study of seed size variation in two invasive shrubs. Ecology,

84: 1434-1440.

Burns, J. H., Pardini, E. A., Schutzenhofer, M. R., Chung, Y. A., Seidler, K. J. &

Knight, T. M. 2013. Greater sexual reproduction contributes to differences in

demography of invasive plants and their noninvasive relatives. Ecology, 94:

995-1004.

Cadotte, M. W. & Lovett-Doust, J. 2001. Ecological and taxonomic differences between

native and introduced plants of southwestern Ontario. Ecoscience, 8: 230-238.

Callaway, R. M. & Ridenour, W. M. 2004. Novel weapons: invasive success and the

evolution of increased competitive ability. Frontiers in Ecology and the

Environment, 2: 436-443.

Catford, J. a., Jansson, R. & Nilsson, C. 2009. Reducing redundancy in invasion

ecology by integrating hypotheses into a single theoretical framework. Diversity

and Distributions, 15: 22-40.

Chrobock, T., Kempel, A., Fischer, M. & van Kleunen, M. 2011. Introduction bias:

cultivated alien plant species germinate faster and more abundantly than native

species in Switzerland. Basic and Applied Ecology, 12: 244-250.

Chun, Y. J., van Kleunen, M. & Dawson, W. 2010. The role of enemy release, tolerance

and resistance in plant invasions: linking damage to performance. Ecology

Letters, 13: 937-946.

Colautti, R. I., Grigorovich, I. A. & MacIsaac, H. J. 2004. Is invasion success explained

by the enemy release hypothesis? Ecology Letters, 7: 721-733.

11 Introduction

Colautti, R. I., Grigorovich, I. A. & MacIsaac, H. J. 2006. Propagule pressure: a null

model for biological invasions. Biological Invasions, 8: 1023-1037.

Colautti, R. I. & Richardson, D. M. 2009. Subjectivity and flexibility in invasion

terminology : too much of a good thing ? Biological Invasions: 1225-1229.

Cornelissen, J. H. C., Lavorel, S., Garnier, E., Diaz, S., Buchmann, N., Gurvich, D. E.,

Reich, P. B., ter Steege, H., Morgan, H. D., van der Heijden, M. G. A., et al.

2003. A handbook of protocols for standardised and easy measurement of plant

functional traits worldwide. Australian Journal of Botany, 51: 335-380.

Davidson, A. M., Jennions, M. & Nicotra, A. B. 2011. Do invasive species show higher

phenotypic plasticity than native species and, if so, is it adaptive? A meta-

analysis. Ecology Letters, 14: 419-431.

Davis, M. A., Grime, J. P. & Thompson, K. E. N. 2000. Fluctuating resources in plant

communities : a general theory of invasibility. Journal of Ecology, 88: 528-534.

Daws, M., Hall, J., Flynn, S. & Pritchard, H. W. 2007. Do invasive species have bigger

seeds? Evidence from intra- and inter-specific comparisons. South African

Journal of Botany, 73: 138-143.

Dawson, W., Fischer, M. & van Kleunen, M. 2011. The maximum relative growth rate

of common UK plant species is positively associated with their global

invasiveness. Global Ecology and Biogeography, 20: 299-306.

Elton, C. S. 1958. The Ecology Of Invasions By Animals And Plants. Methue, London,

U.K.

Felker-Quinn, E., Schweitzer, J. a. & Bailey, J. K. 2013. Meta-analysis reveals

evolution in invasive plant species but little support for Evolution of Increased

Competitive Ability (EICA). Ecology and Evolution, 3: 739-751.

12 Introduction

Garcia-Serrano, H., Escarré, J. & Sans, F. X. 2004. Factors that limit the emergence and

establishment of the related aliens Senenecio inaequidens and Senecio

pterophorus and the native Senecio malacitanus in Mediterranean climate.

Canadian Journal of Botany, 82: 1346-1355.

Godoy, O., Valladares, F. & Castro-Díez, P. 2011. Multispecies comparison reveals that

invasive and native plants differ in their traits but not in their plasticity.

Functional Ecology, 25: 1248-1259.

Grigulis, K., Sheppard, A. W., Ash, J. E. & Groves, R. H. 2001. The comparative

demography of the pasture weed plantagineum between its native and

invaded ranges. Journal of Applied Ecology, 38: 281-290.

Grotkopp, E. & Rejmánek, M. 2007. High seedling relative growth rate and specific leaf

area are traits of invasive species: Philogenetically independent contrasts of

woody angiosperms. American Journal of Botany, 94: 526-532.

Hawkes, C. V. 2007. Are invaders moving targets? The generality and persistence of

advantages in size, reproduction, and enemy release in invasive plant species

with time since introduction. American Naturalist, 170: 832-843.

Hedges, L. V., Gurevitch, J. & Curtis, P. S. 1999. The meta-analysis of response ratios

in Ecology. Ecology, 80: 1150-1156.

Henery, M. L. & Westoby, M. 2001. Seed mass and seed nutrient content as predictors

of seed output variation between species. Oikos, 92: 479-490.

Hierro, J. L., Eren, Ö., Villarreal, D. & Chiuffo, M. C. 2013. Non-native conditions

favor non-native populations of invasive plant: demographic consequences of

seed size variation? Oikos, 122: 583-590.

13 Introduction

Hierro, J. L., Maron, J. L. & Callaway, R. M. 2005. A biogeographical approach to

plant invasions : the importance of studying exotics in their introduced and

native range. Journal of Ecology: 5-15.

Higgins, J. P. T., Green, S. & Collaboration, C. 2008. Cochrane handbook for

systematic reviews of interventions.

Jongejans, E., Shea, K., Skarpaas, O., Kelly, D., Sheppard, A. W. & Woodburn, T. L.

2008. Dispersal and demography contributions to population spread of Carduus

nutans in its native and invaded ranges. Journal of Ecology, 96: 687-697.

Kattge, J., Díaz, S., Lavorel, S., Prentice, I. C., Leadley, P., Bönisch, G., Garnier, E.,

Westoby, M., Reich, P. B., Wright, I. J., et al. 2011. TRY - a global database of

plant traits. Global Change Biology, 17: 2905-2935.

Keane, R. M. & Crawley, M. J. 2002. Exotic plant invasions and the enemy release

hypothesis. Trends in Ecology & Evolution, 17: 164-170.

Lake, J. C. & Leishman, M. R. 2004. Invasion success of exotic plants in natural

ecosystems: the role of disturbance, plant attributes and freedom from

herbivores. Biological Conservation, 117: 215-226.

Levine, J. M., Adler, P. B. & Yelenik, S. G. 2004. A meta-analysis of biotic resistance

to exotic plant invasions. Ecology Letters, 7: 975-989.

Liu, H. & Stiling, P. 2006. Testing the enemy release hypothesis: a review and meta-

analysis. Biological Invasions, 8: 1535-1545.

Lodge, D. M. 1993. Biological Invasions: lessons for Ecology. Trends in Ecology &

Evolution, 8: 133-137.

Lonsdale, W. M. 1999. Global Patterns of Plant Invasions and the Concept of

Invasibility. Ecology, 80: 1522-1536.

14 Introduction

Maron, J. L., Vilà, M., Bommarco, R., Elmendorf, S. & Beardsley, P. 2004. Rapid

evolution of an invasive plant. Ecological Monographs, 74: 261-280.

Mason, R. A. B., Cooke, J., Moles, A. T. & Leishman, M. R. 2008. Reproductive output

of invasive versus native plants. Global Ecology and Biogeography, 17: 633-

640.

Moles, A. T., Warton, D. I., Warman, L., Swenson, N. G., Laffan, S. W., Zanne, A. E.,

Pitman, A., Hemmings, F. a. & Leishman, M. R. 2009. Global patterns in plant

height. Journal of Ecology, 97: 923-932.

Moles, A. T. & Westoby, M. 2006. Seed size and plant strategy across the whole life

cycle. Oikos, 113: 91-105.

Nakagawa, S. & Poulin, R. 2012. Meta-analytic insights into evolutionary ecology: an

introduction and synthesis. Evolutionary Ecology, 26: 1085-1099.

Ordonez, A., Wright, I. J. & Olff, H. 2010. Functional differences between native and

alien species: a global-scale comparison. Functional Ecology, 24: 1353-1361.

Palacio-López, K. & Gianoli, E. 2011. Invasive plants do not display greater phenotypic

plasticity than their native or non-invasive counterparts: a meta-analysis. Oikos,

120: 1393-1401.

Parker, J. D., Torchin, M. E., Hufbauer, R. A., Lemoine, N. P., Alba, C., Blumenthal, D.

M., Bossdorf, O., Byers, J. E., Dunn, A. M., Heckman, R. W., et al. 2013. Do

invasive species perform better in their new ranges? Ecology, 94: 985-994.

Prinzing, A., Durka, W., Klotz, S. & Brandl, R. 2002. Which species become aliens?

Evolutionary Ecology Research, 4: 385-405.

Pugnaire, F. & Valladares, F. 2007. Functional plant ecology. CRC Press, Florida,

USA.

15 Introduction

Pyšek, P., Prach, K. & Smilauer, P. 1994. Relating invasion success to plant traits: an

analysis of the czech alien flora. 39-60.

Pyšek, P. & Richardson, D. M. 2007. Traits associated with invasiveness in alien plants:

where do we stand? In: W. Nentwig (ed.), pp. 97-125. Biological Invasions,

Ecological Studies Vol. 193, pp. 97-125. Springer-Verlag, Berlin.

Rejmánek, M. 1996. A theory of seed plant invasiveness: the first sketch. Biological

Conservation, 78: 171-181.

Richardson, D. M. & Pyšek, P. 2006. Plant invasions: merging the concepts of species

invasiveness and community invasibility. Progress in Physical Geography, 30:

409-431.

Richardson, D. M., Pyšek, P., Rejmánek, M., Barbour, M. G., Dane, F. & West, C. J.

2000. Naturalization and invasion of alien plants: concepts and definitions.

Diversity and Distributions, 6: 93-107.

Sakai, A. K., Allendorf, F. W., Holt, J. S., Lodge, D. M., Molofsky, J., With, K. A.,

Baughman, S., Cabin, R. J., Cohen, J. E., Ellstrand, N. C., et al. 2001. The

population biology of invasive species. Annual Review of Ecology and

Systematics, 32: 305-332.

Sax, D. F., Stachowicz, J. J., Brown, J. H., Bruno, J. F., Dawson, M. N., Gaines, S. D.,

Grosberg, R. K., Hastings, A., Holt, R. D., Mayfield, M. M., et al. 2007.

Ecological and evolutionary insights from species invasions. Trends in Ecology

& Evolution, 22.

Schlaepfer, D. R., Glättli, M., Fischer, M. & van Kleunen, M. 2010. A multi-species

experiment in their native range indicates pre-adaptation of invasive alien plant

species. New Phytologist, 185: 1087-1099.

16 Introduction

Silvertown, J., Franco, M., Pisanty, I. & Mendoza, A. 1993. Comparative plant

demography- relative importance components of life-cycle components to the

finite rate of increase in woody and herbaceous perennials. Journal of Ecology,

81: 465-476.

Simberloff, D. 2006. Invasional meltdown 6 years later: important phenomenon ,

unfortunate metaphor, or both? Ecology Letters: 912-919.

Simberloff, D. & von Holle, B. 1999. Positive interactions of nonindigenous species:

invasional meltdown? Biological Invasions, 1: 21-32.

Thébaud, C. & Simberloff, D. 2001. Are plants really larger in their introduced ranges?

American Naturalist, 157: 231-236.

Thompson, K. & Davis, M. A. 2011. Why research on traits of invasive plants tells us

very little. Trends in Ecology & Evolution, 26: 155-156.

Thompson, K., Hodgson, J. G. & Rich, T. C. G. 1995. Native and alien invasive plants:

more of the same? Ecography, 18: 390-402. van Kleunen, M., Dawson, W. & Dostal, P. 2011. Research on invasive-plant traits tells

us a lot. Trends in Ecology & Evolution, 26: 317; author reply 319-317; author

reply 319. van Kleunen, M., Dawson, W., Schlaepfer, D., Jeschke, J. M. & Fischer, M. 2010. Are

invaders different ? A conceptual framework of comparative approaches for

assessing determinants of invasiveness. Ecology Letters, 13: 947-958. van Kleunen, M. & Johnson, S. D. 2007. South African Iridaceae with rapid and profuse

seedling emergence are more likely to become naturalized in other regions.

Journal of Ecology, 95: 674-681. van Kleunen, M., Weber, E. & Fischer, M. 2010. A meta-analysis of trait differences

between invasive and non-invasive plant species. Ecology Letters, 13: 235-245.

17 Introduction

Vetter, D., Rucker, G. & Storch, I. 2013. Meta-analysis: A need for well-defined usage

in ecology and conservation biology. Ecosphere, 4: 1-24.

Vilà, M., Maron, J. L. & Marco, L. 2005. Evidence for the enemy release hypothesis in

Hypericum perforatum. Oecologia, 142: 474-479.

Westoby, M., Falster, D. S., Moles, A. T., Vesk, P. A. & Wright, I. J. 2002. Plant

ecological strategies: some leading dimensions of variation between species.

Annual Review of Ecology and Systematics, 33: 125-159.

Westoby, M., Jurado, E. & Leishman, M. 1992. Comparative evolutionary ecology of

seed size. Trends in Ecology & Evolution, 7: 368-372.

Williamson, M. 1999. Invasions. Ecography, 5: 5-12.

Williamson, M. H. & Fitter, A. 1996. The varying success of invaders. Ecology, 77:

1661-1666.

18 Are introduced species better dispersers?

Chapter one

Are introduced species better dispersers than native

species? A global comparative study of seed dispersal

distance

Habacuc Flores-Moreno, Fiona J. Thomson, David I. Warton & Angela T. Moles

PLoS ONE (2013) 8(6): e68541. doi:10.1371/journal.pone.0068541

The study was conceived and designed by HFM and ATM, HFM and FJT performed

the experiments and collated the data. HFM, DIW and ATM analysed the data. DIW

provided analytical tools. The manuscript writing was led by HFM.

This chapter has been modified from the original paper specifically for this thesis.

Abstract

We provide the first global test of the idea that introduced species have greater seed dispersal distances than do native species, using data for 51 introduced and 360 native species from the global literature. Counter to our expectations, there was no significant difference in mean or maximum dispersal distance between introduced and native species. Next, we asked whether differences in dispersal distance might have been obscured by differences in seed mass, plant height and dispersal syndrome, all traits that

19 Are introduced species better dispersers?

affect dispersal distance and which can differ between native and introduced species.

When we included all three variables in the model, there was no clear difference in dispersal distance between introduced and native species. These results remained consistent when we performed analyses including a random effect for site. Analyses also showed that the lack of a significant difference in dispersal distance was not due to differences in biome, taxonomic composition, growth form, nitrogen fixation, our inclusion of non-invasive introduced species, or our exclusion of species with human- assisted dispersal. Thus, if introduced species do have higher spread rates, it seems likely that these are driven by differences in post-dispersal processes such as germination, seedling survival, and survival to reproduction.

Introduction

It has often been suggested that introduced and/or invasive species have greater natural dispersal abilities than do native or less invasive species [1-4]. For instance Murray and

Phillips [1] state that “naturalized invasive species that spread by seed possess enhanced strategies for seed dispersal that are either absent or at least not as prevalent in naturalized species that are not invasive”. Thompson and Davis [5] state that research on traits of invasive plants “has revealed that, when compared with natives or non-invasive aliens, invasive aliens…produce more seeds that are better dispersed…”, and Ordonez and Olff [4] state that “analyses comparing regional and global species pools of natives and aliens have found that aliens …. produce more seeds that are better dispersed…”. The idea that introduced species disperse their seeds better than natives species underpins much of our understanding of the dynamics of introduced species, including rates of spread [6-8], range sizes [6,7,9] and the ability of introduced species to take advantage of colonization opportunities arising from disturbance and/or an

20 Are introduced species better dispersers?

increase in resource levels [10-13]. In this study, we provide the first general test of the fundamental idea that introduced plant species achieve greater dispersal distances under natural conditions than do native species.

The empirical evidence underlying the idea that introduced species have longer seed dispersal distances than do native species is mixed, based on data for relatively few species, and often relies on indirect measures of dispersal capacity (e.g. size of dispersal structures, terminal velocity or buoyancy). For example, some studies have found that introduced species are more likely to have long-distance dispersal vectors, like humans

[14], wind [15] and vertebrates [15]. However other studies have found the contrary, with introduced species less likely to be dispersed by vertebrates [16], no difference in the proportion of expanding natives and invasive introduced species that are wind dispersed [17] or that introduced species are dispersed by human vectors, but not by wind or animal vectors [18]. The few studies that have compared actual dispersal distances between pairs or small sets of introduced and co-occurring native species have also found conflicting results, reporting introduced species to have greater seed dispersal distances than native species [19], or no significant difference between introduced and native species [20,21].

Previous attempts to generalize across local/taxon specific studies have used vote counting [qualitative comparison of the number of studies showing significant and non- significant differences; 3,22], finding a weak trend towards greater seed dispersal capacity for introduced invasive species. Unfortunately, the number of studies included in each review was less than eight, no statistical comparison of the dispersal advantage of introduced over native species was done, and some of the studies analyzed in these

21 Are introduced species better dispersers?

reviews inferred dispersal ability from dispersal related traits. Thus, a global, quantitative test of the hypothesis that introduced species have greater seed dispersal distances than do native species was urgently needed. This was our main aim.

Our first step was to run a simple comparison of the dispersal distances achieved by introduced and native species. This comparison will help us to understand the role of dispersal distance in the generation of spatial patterns and species composition in plant communities. For example, differences in dispersal can contribute to differences in relative abundance and species richness between plant communities [23]. Differences in seed dispersal distances can also lead to differences in post-dispersal processes such as seed mortality, germination and seedling survival [6,24]. Plants with short seed dispersal distances are thought to be restricted by greater density-dependent mortality, while plants dispersing further are more restricted by the lack of reproductive partners or the lack of suitable microhabitats [25]. Thus, differences in seed dispersal distance can lead to differences in selective pressure and therefore differences in life histories between introduced and native species.

Seed mass, plant height and dispersal syndrome are crucial ecological traits [26-32] that have been shown to affect the dispersal of plant species [24,33], and which sometimes differ between native and introduced species [14-16, but see 17,18,34-39, and 40].

Differences in plant height, seed mass and/or dispersal syndrome could mask or artificially generate a difference in dispersal distance between native and introduced species. We therefore asked a) whether there were significant differences in seed mass, plant height and dispersal syndrome between the native and introduced species in our dataset, and b) whether there were differences in dispersal distance between introduced

22 Are introduced species better dispersers?

and native plants once we had accounted for plant height, seed mass and dispersal syndrome. Determining whether the differences in dispersal distance remain significant after accounting for these factors will allow us to determine whether any greater dispersal ability of introduced species arises as part of a coordinated suite of life history traits, or whether the dispersal abilities of introduced species might result from selection on aspects of seed morphology, such as wing/pappus size. Our results will also further our understanding of the suite of traits that distinguish introduced from native species.

In summary, the hypotheses we address are:

1) Introduced species will have greater dispersal distances than do native species.

2) Introduced species will have greater dispersal distances than do native species once plant height, seed mass and dispersal syndrome have been accounted for.

Materials and Methods

Ethics statement

All data in this study were extracted from published sources, hence no permission or approval for obtaining the data was required.

Data collection

We began with the seed dispersal database generated by Thomson et al. [33]. These data were primarily for native species, and there is no similar database for introduced species. We therefore performed a search on ISI Web of Knowledge for papers published between 1906 and 2010 with information on seed dispersal distance of introduced plants. The search terms we used were „seed‟ + „dispersal distance‟ or „seed dispersal‟ + „distance‟, and „dispersal kernel‟, „dispersal curve‟ or „seed shadow‟

23 Are introduced species better dispersers?

restricted by the terms „weed$‟, „introduced‟, „invasive‟,‟ non-invasive‟, „naturaliz*‟,

„alien‟, „non-native‟, and „noxious‟. We also searched for relevant papers in the reference lists of focal papers. We only included papers that presented observational or experimental information on the mean and/or maximum dispersal distance of plant species under field conditions. This included studies that used seed traps, tracked individual seeds, marked and recaptured seeds and estimated dispersal distances based on tracking vectors and calculating gut or fur retention times. We excluded all studies that estimated seed dispersal distances from seed size, including mass and shape, or dispersal syndrome because these variables were included in the analyses. We excluded studies that used artificial or unrealistic conditions, such as wind tunnels, artificial fur or artificial seeds. Studies that calculated seed dispersal distance from buoyancy tests or terminal velocity tests were also excluded. Studies that estimated dispersal distances based on spatial population mathematical models and inverse modeling of seedling distance to nearest adult or mother plant were excluded because they are influenced by post-dispersal processes such as germination success, seed predation and seedling predation. Studies with less than ten replicates for a given species for either mean or maximum dispersal distance were also excluded. When observational and experimental information were both included in the study, we preferentially used observational information.

Information on species‟ status (native/introduced) was extracted from the same papers as dispersal distance data. Information on dispersal distances were extracted from three sources (in diminishing order of preference): 1) tables, 2) main text or 3) graphs using

DatathiefIII [41]. Maximum and mean dispersal distances were used instead of percentiles since these were the most common measures throughout the literature.

24 Are introduced species better dispersers?

When possible, information on seed mass and maximum plant height or seed release height was extracted from the same papers as dispersal distance data. Information on seed mass and plant height were extracted from the same three sources as dispersal distance data: 1) tables, 2) main text or 3) graphs using DatathiefIII [41]. Otherwise, information on seed mass and plant height were taken from Moles et al. [26,42], the

Royal Botanic Gardens Kew‟s Seed Information Database [43] or Mason et al. [40]. For each species we used maximum plant height where possible, but where these data were not available we used the maximum recorded mean plant height. We extracted information on dispersal syndrome from the same papers as dispersal distance data.

Dispersal syndrome information was only extracted when the paper explicitly stated that the dispersal distance data were associated with a given dispersal syndrome. Species were initially grouped into four dispersal syndromes: wind, water, animal and unassisted. However, our sample size for water dispersal was very low, and since mean dispersal distance of water and wind dispersed species do not significantly differ (P =

0.70) they were treated as one category (water/wind syndrome) to increase our statistical power. In the case of maximum dispersal distance water-dispersed species were excluded because of scarce data (n = 8). Therefore only three categories were used in final analyses. Data for species with more than one dispersal syndrome were included as separate data points for each syndrome.

In total our database contained information on 411 species from 92 families, including data for 360 species in their home range and 51 species in their introduced range. Of the

51 introduced species in our study, five (12 %) were classified as naturalized [species that establish self-replacing populations without range expansion; sensu 44] and 46 (88

25 Are introduced species better dispersers?

%) as invasive [introduced species with rapid population increase and range expansion; sensu 44]. Introduced species represented 12.4 % of our seed dispersal distance database. Although this is a modest proportion of the whole dataset, it follows the trend reported by Vitousek et al. [45], where on average introduced species represent 8.3 % of large continental area floras (e.g. , Tropical , Chile) and 13.3 % of smaller continental area floras (e.g. Egypt, Queensland, Texas).

Dispersal distance, seed mass and plant height data were log10-transformed before analysis.

Data analyses

We began by asking whether there were differences in dispersal distance, plant height, seed mass and dispersal syndrome between native and introduced species. We compared the dispersal distances, heights and seed masses of native and introduced species with

Student‟s t-tests, assuming unequal variance. To compare dispersal syndrome between native and introduced species, we used a contingency table. Analyses for dispersal syndrome were performed both with all available data, and excluding water-dispersed species for which we had a very small sample size for introduced species (n = 4).

Results were consistent, and for brevity, we present only the more robust analysis based on the three well-replicated dispersal syndromes.

Next, we asked whether there were differences in mean and maximum dispersal distance when accounting for plant height, seed mass or dispersal syndrome individually. For seed mass and plant height, we used ANCOVAs in which introduced/native status was our categorical variable, seed mass or plant height our

26 Are introduced species better dispersers?

covariates, and mean or maximum dispersal distance was the dependent variable. We began by confirming that our data fulfilled the homogeneity of variance assumption for

ANCOVA (Figure S1). ANCOVAs included terms for species‟ status, a trait covariate

(plant height or seed mass) and their interaction (status × trait). A significant effect of species‟ status would show different average dispersal distances between status, for a given value of the trait covariate. A significant difference in the trait covariate (seed mass or plant height) would show a slope ≠ 0, for species of a given status. A significant effect of the interaction between status and the trait variable would show a difference in the relationship between dispersal distance and the trait covariate, between introduced and native species. That is, introduced and native species would have different slopes and intercepts at the same time. To test whether introduced and native species differ in dispersal distance once the effect of dispersal syndrome had been accounted for we ran a linear model where species‟ status and dispersal syndrome were our predictor variables and mean or maximum dispersal distances were our response variables.

We ran a linear model to test whether dispersal distance differs between species as an effect of their native or introduced status once dispersal syndrome, seed mass and plant height had been accounted for. For these linear models our predictor variables were species‟ status (native or introduced), dispersal syndrome, seed mass and plant height, and our dependent variables were maximum and mean dispersal distance.

Data considerations

We did not have enough data (and thus degrees of freedom) to explicitly control for every possible correlate of dispersal distance. We therefore selected a few particularly relevant traits (seed mass, plant height and dispersal syndrome) to include in the main

27 Are introduced species better dispersers?

analyses. However, there are other ecological variables that could have an important effect on dispersal distance. We therefore concluded with a series of analyses that explored the potential effects of site to site variation, biome, , growth form, ability to fix nitrogen, introduced species‟ level of invasiveness (naturalized vs. invasive species) and human assisted dispersal.

The majority of our analyses were run as linear models. However, some datapoints came from the same sites, and so are not fully independent. To address this, we ran models including a random effect for site. These analyses cannot be run with missing values, so first we generated and reanalyze a subset of data with no missing values. The results were broadly consistent with previous results (Table S3). Then, we reanalyze the subset of data with no missing values including a random effect for site (Supporting

Information S1).Information on the number of missing values for seed mass, plant height and dispersal syndrome data are available in Table S4.

We tested whether the dispersal distance of introduced and native species differed once the effect of biome (tropical forest, temperate forest, grassland, shrubland, woodland and other; Table S1) had been accounted for using a linear model where biome and species‟ status (introduced and native) where our predictor variables and mean or maximum dispersal distance were our dependent variables.

Our next step was to calculate the taxonomic distribution of our study species (Table

S1). We used a Yates‟ Chi-square because of the high proportion of cells in the analysis with expected values less than five [46].We did not perform phylogenetic contrast analyses because neither species‟ status, nor dispersal distance are heritable traits.

28 Are introduced species better dispersers?

Although certain taxa are more likely to be introduced [47,48], a plant doesn‟t evolve to be introduced. Dispersal distance is affected by heritable traits such as seed mass, plant height and dispersal syndrome (which we consider in our analyses), but dispersal distance is also affected by non-heritable factors such as the characteristics of the surrounding landscape, wind conditions, the availability of dispersers, and a good deal of chance; [32,49]. As phylogenetic analyses are explicitly evolutionary (for instance,

Felsenstein‟s 1985 [50] method assumes that traits evolve under Brownian motion along branches), phylogenetic analyses are not appropriate for our data.

We used contingency tables to test whether there was a difference in the proportion of introduced and native species with a given growth form (woody vs. non-woody) and with or without nitrogen-fixing capacity.

Many of the predictions about the differences between introduced and native species, and the studies that test these predictions are phrased broadly to include all introduced species [4,35,38,51,52]. However, other predictions are about the difference between native species and invasive introduced species. Only a small proportion of introduced species become invasive [53]. Our main analyses allow us to address the broader question of whether introduced species differ functionally from native species [e.g.

35,54,55] and if they do, to identify which traits or characteristics are associated with species that have become established in new environment [e.g. 4,56,57]. However, we ran ANOVA tests comparing native species to a subset of invasive introduced species (n

= 46), to determine whether the patterns we observed across the full dataset hold up when non-invasive introduced species were excluded. Our predictor variable was

29 Are introduced species better dispersers?

species status (native or introduced) and our dependent variables were mean or maximum dispersal distance.

Transport of seeds by humans (for example, by harvesters or mowing machines) was excluded from the main analyses because it is clearly a different process to natural seed dispersal. However, to test whether the inclusion of human-transported species had any effect on our results we ran ANOVA tests including human-dispersed species (two native and five introduced). In these analyses our predictor variable was species‟ status and the dependent variable was mean or maximum dispersal distance.

All analyses were performed in R [58], with species as the replicates. We report partial

R2s throughout. Species‟ status (introduced vs. native) is a predictor variable in analyses, rather than the dependent. That is, our analyses ask whether native and introduced species differ in dispersal distance, rather than using dispersal distance to predict whether a species will be introduced or not.

30 Are introduced species better dispersers?

Results

Contrary to our expectations, we found no significant difference between native and introduced species‟ mean (P = 0.18, Fig. 1A) or maximum seed dispersal distances (P =

0.43; Fig. 1B).

Figure 1. Comparison of native and introduced species’ dispersal distances, plant height and seed mass. Black dashed lines represent mean values. The boxes represent the 25th, 50th and 75th percentiles. Whiskers represent the 10th and 90th percentiles, outliers are represented as points. Sample sizes are number of species

Native plant species were approximately twice as tall as were introduced species (P =

0.01, Fig. 1C) and had seeds three times bigger (P = 0.009, Fig. 1D) than did introduced plant species. The proportion of introduced and native species using a given dispersal

31 Are introduced species better dispersers?

syndrome was also significantly different (Χ2 = 42.43, d.f. = 2, P < 0.0001). There were significantly more introduced (53.4%) than native (21.7%) species with wind- or water- dispersed seeds. There were also significantly more introduced (29.3%) than native species (16.1%) with unassisted dispersal. Finally, there was a significantly higher proportion of native species (62.2%) than introduced species (17.2%) with animal dispersal (Table S1).

As expected, seed mass, plant height and dispersal syndrome all affected dispersal distance. Seed mass had a significant positive main effect on both mean (P < 0.0001) and maximum (P = 0.005) dispersal distance, and dispersal distance increased with increasing plant height in all analyses (P < 0.0001). On average, animal-dispersed species had the highest mean (33.4 m) and maximum dispersal distance (51.5 m), followed by water/wind-dispersed species for mean dispersal distance (3.1 m) and wind-dispersed species for maximum dispersal distance (22.04 m). Species with unassisted dispersal had the lowest dispersal distance of all syndromes for both mean

(0.7 m) and maximum (2.5 m) dispersal distance.

When we accounted for the effect of one plant trait (seed mass, plant height or dispersal syndrome), we tended to find significant interactions between species status and the plant trait under consideration (Figure S1 and Table S2). For mean and maximum dispersal distance we found a significant interaction between plant height and species status (P = 0.03 and < 0.0001 respectively). That is, differences in dispersal distance between introduced and native species vary with plant height. There were no significant differences between introduced and native species‟ mean dispersal distance after accounting for the effect of seed mass (P = 0.43) or dispersal syndrome (P = 0.48).

32 Are introduced species better dispersers?

However, for maximum dispersal distance there were significant interactions between seed mass and species‟ status (P = 0.05) and between dispersal syndrome and species‟ status (P = 0.003). That is, the difference in maximum dispersal distance between introduced and native species varies with seed mass or dispersal syndrome.

To determine whether species‟ status affected dispersal distance after accounting for variation due to seed mass, plant height, and dispersal syndrome simultaneously, we constructed a model including terms for status, dispersal syndrome, plant height and seed mass and interactions. For mean dispersal distance, none of the five interactions between species‟ status seed mass, plant height and/or dispersal syndrome were significant (P > 0.18; Table 1), and the main effect of status was not significant (P =

0.67; Table 1). That is, once we have accounted for the effect of seed mass, plant height and dispersal syndrome, there is no significant difference in mean seed dispersal distance between native and introduced species. Interestingly, height (R2 = 0.23) explained almost six times as much variation as did dispersal syndrome (R2 = 0.04), which was the next best predictor of mean dispersal distance, and seed mass (R2 = 0.02) and species‟ status (R2 < 0.001) made much smaller contributions to the predictive power of the model. The overall model explained 63 % of the variation in mean dispersal distance, a remarkable outcome given that our data come from a range of taxa in a broad range of ecosystems worldwide.

33 Are introduced species better dispersers?

Table 1. Effect of species’ status (native vs. introduced), seed mass, plant height and dispersal syndrome (animal, unassisted, water/wind), and their interactions on mean dispersal distance.

Term Sum of Squares F ratio P

Status 0.08 0.18 0.67

Seed mass 2.85 6.61 0.01

Plant height 30.68 71.33 < 0.0001

Dispersal syndrome 5.29 6.15 0.003

Status × Dispersal syndrome 0.15 0.17 0.85

Status × Seed mass 0.27 0.63 0.43

Status × Plant height 0.12 0.27 0.60

Status × Plant height × Dispersal syndrome 2.68 1.56 0.19

Status × Seed mass × Dispersal syndrome 2.11 1.23 0.30

We next constructed a model for maximum dispersal distance including terms for status, dispersal syndrome, plant height and seed mass and interactions. There were two significant three-way interactions, and two significant two-way interactions, all including status (Table 2). That is, the effect of status differs according to height, seed mass and dispersal syndrome. As in the model for mean dispersal distance, the main effect of species‟ status was not significant (P = 0.23). However, most statisticians advise against interpreting main effects in the presence of significant higher order interactions [59].

34 Are introduced species better dispersers?

Table 2. Effect of species’ status (native vs. introduced), seed mass, plant height and dispersal syndrome (animal, unassisted, water/wind), and their interactions on maximum dispersal distance.

Term Sum of F ratio P

Squares

Status 0.58 1.43 0.23

Seed mass 1.52 3.73 0.06

Plant height 3.35 8.21 0.005

Dispersal syndrome 23.09 28.31 < 0.0001

Status × Dispersal syndrome 0.25 0.31 0.74

Status × Seed mass 4.37 10.72 0.001

Status × Plant height 3.60 8.82 0.003

Status × Plant height × Dispersal syndrome 8.46 5.19 <0.001

Status × Seed mass × Dispersal syndrome 8.08 4.95 <0.001

Data considerations

1. Site to site variation.

Analyses including a random effect for site (Supporting Information S1) were broadly consistent with previous analyses. That is, the fact that species occurring at the same site are not fully independent is not the reason for a lack of a significant difference in the dispersal distance of introduced and native species.

2. Biome.

The main effect of species status (native vs. introduced) was not significant for either mean or maximum dispersal distance once biome had been accounted for (P = 0.07 and

35 Are introduced species better dispersers?

0.29 respectively). There was a significant interaction between biome and species‟ status for maximum dispersal distance (P = 0.0007). That is, the maximum dispersal distance achieved by introduced and native species varies by biome. However, there is not a consistent difference in maximum dispersal distance between introduced and native species after accounting for the effect of biome.

3. Taxonomy.

Although there were some modest differences in the relative representation of different taxonomic groups, between native and introduced species (Table S1), these differences were not significant (Yates‟ χ2 = 14.71, d.f. = 9, P = 0.10).

4. Growth form.

A significantly higher proportion of native species than introduced species had a woody growth form (Χ2 = 18.55, d.f. = 1, P < 0.0001; Table S1). However, when we ran a linear model with species‟ status, plant height, seed mass and growth form as predictor variables, neither growth form (P > 0.20 ) nor species‟ status (P > 0.66) had a significant effect on mean or maximum dispersal distance.

5. Ability to fix nitrogen.

There was no significant difference in the proportion of introduced and native species that are N-fixers (Yates Χ2 = 0.09, d.f. = 1, P = 0.77; Table S1).

6. Naturalized vs. invasive species.

An analysis of the differences between species with varying levels of invasiveness is outside of the scope of our present study. However, only five of our introduced species

36 Are introduced species better dispersers?

were not invasive. Excluding these five species from analysis did not qualitatively affect our results. There was no significant difference in the mean (P = 0.14) or maximum (P

= 0.63) seed dispersal distance when the non-invasive species were excluded from analysis.

7. Human-assisted dispersal.

There was no significant difference between the mean (P = 0.23) or maximum (P =

0.25) seed dispersal distance of introduced and native species when the seven human- dispersed species were included in the analysis. That is, including species with human- assisted dispersal did not qualitatively affect our results.

In summary, the lack of a significant difference in average dispersal distance between native and introduced species is not due to a relationship being obscured by differences in the biomes from which the data were taken, the taxonomic composition of the datasets, the growth form of the species, or differences in the proportion of native vs. introduced species that were able to fix nitrogen. We can also exclude the possibility that our inclusion of species with different levels of invasiveness, our exclusion of species with human-assisted dispersal, or our treatment of replicate species from the same sites has masked a significant difference in seed dispersal distance between introduced and native species.

Discussion

While there are some invasive plants that have truly spectacular rates of spread across the landscape (e.g. Pueraria lobata in the or Abutilon theophrasti in

Europe), our data show that there is no overall difference in dispersal distances achieved

37 Are introduced species better dispersers?

by native and introduced species. This pattern holds after accounting for variation due to traits such as seed mass, plant height and dispersal syndrome. Our findings challenge traditional assumptions about the importance of dispersal distance for the spread of introduced species [2,6,17,60], and have two implications for management. First, there is no fundamental reason to expect native species to be more limited in their natural dispersal ability when responding to global change than are introduced species. Second, we need to remember that introduced species, like native species, have a range of traits, and do not all have exceptional dispersal abilities.

There are often differences in dispersal-related traits between native and introduced/invasive species [35,61-63]. It is clearly important to ask whether there are differences in the dispersal distances achieved by native vs. introduced species when all else is equal (that is, when comparing similar types of species). However, the important variable determining rates of spread of introduced and native species is not dispersal distance after accounting for other variables, but simply how far the seeds of each group of species travel. That is, if introduced species did differ in a trait with native species, and this helped their seeds to disperse greater distances, the end results would still be a greater rate of spread for introduced species.

Although seed dispersal is an important determinant of plant distributions, getting there is just part of the struggle, and seed arrival does not necessarily translate to recruitment

[6]. Thompson and Davis [5] state that distinction between plant winners or losers

“often owes rather little to native or alien status”. In the case of dispersal distance this is true. Species‟ status explained less than 1 % of the variation in both mean and maximum dispersal distance. To fully understand introduced species‟ high spread rates

38 Are introduced species better dispersers?

and rapid colonization we need to consider other factors that could favor rapid spread; these may include short times to reproduction, high post dispersal survival, high germination, human mediated processes or propagule pressure [52,64-66]. In particular, association to humans has been suggested to be key to the long distance dispersal of invasive species [67] and in general to be a strong determinant of species‟ invasiveness

[63,68]. Clearly, the contribution of human-mediated dispersal and post-dispersal mechanisms is an area of invasion biology where future research is urgently needed in order to understand the patterns and processes that govern the spread of introduced species.

Not all introduced species are invasive. We wondered whether over representation of non-invasive introduced species [naturalized species sensu Richardson; 44] in our data set could have obscured differences in dispersal distance between introduced and native species. However, of the 51 introduced species included in our study, only five were classified as naturalized by the primary studies or environmental agencies of the countries where the studies were done, while 46 were classified as invasive. Thus, a lack of invasive species is not responsible for the lack of a significant difference between dispersal distance in native and introduced species. If anything, the predominance of invasive introduced species in our dataset would be expected to bias our study towards finding a significant difference in dispersal ability between introduced and native species, particularly since ecologists may be more likely to study seed dispersal on introduced species that have relatively rapid rates of spread.

Although we have by far the largest dataset ever compiled on seed dispersal distance, introduced species had a low sample size in mean and maximum dispersal distance (n =

39 Are introduced species better dispersers?

41 and 49, respectively) compared to native species (n = 248 and 288 for mean and maximum dispersal distance, respectively). It is possible that the low sample size in introduced species could have contributed to the non-significant difference in dispersal distance between introduced and native species (Figure 1 A and B). However a power test [59] revealed that to find a 5% difference (with a 95% accuracy) between introduced and native species mean and maximum dispersal distance we only would have needed 39 and 32 replicates respectively. That is, any difference in dispersal distance between introduced and native species would have to be very small in order to not have been detected in our main comparisons.

We still have a great deal to learn about the factors underlying the success of introduced species, and our study highlights several important questions for the future. First, higher recruitment success could play a major role in facilitating the establishment and rapid population growth rate of introduced species [3,52,69]. Thus, determining whether introduced species have higher recruitment success under natural conditions is a promising direction. Second, species with different levels of invasiveness might differ in their traits and life history characteristics [70,71]. Quantifying the differences between species with different levels of invasiveness could help us identify which traits or trait values are most strongly associated with the success of invasive species [70,71].

Finally, using formal path analyses we could determine the extent to which observed differences in fundamental ecological traits between introduced and native species actually translate to differences in recruitment and rates of spread. That is, we could test the adaptive nature of the morphological and functional traits associated with introduced and/or invasive species.

40 Are introduced species better dispersers?

The idea that introduced species, particularly invasive species, are better dispersers than are native species was based on the observation of high spread rates, high population growth rates and broad differences in dispersal related traits that theoretically would give introduced species great advantage in achieving high dispersal distances [7-9,40].

However, our data are not consistent with this idea. Our findings advance our knowledge of the ecological characteristics and mechanisms that underlie the spatiotemporal dynamics of introduced species. These concepts are not only central to understanding the ecology of introduced species, but are also central to their management.

Acknowledgments

H.F.M. wishes to thank Rhiannon Dalrymple, José Antonio Flores Díaz and Ken

Thompson for comments on previous versions of the manuscript. H.F.M. was supported by a scholarship from the Evolution & Ecology Research Centre at UNSW. A.T.M. was supported by funding from the Australian Research Council (DP 0984222). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1. Murray BR, Phillips ML (2010) Investment in seed dispersal structures is linked to

invasiveness in exotic plant species of south-eastern . Biol Invasions

12: 2265-2275.

2. Jongejans E, Shea K, Skarpaas O, Kelly D, Sheppard AW, et al. (2008) Dispersal and

demography contributions to population spread of Carduus nutans in its native

and invaded ranges. J Ecol 96: 687-697.

41 Are introduced species better dispersers?

3. Pyšek P, Richardson DM (2007) Traits associated with invasiveness in alien plants:

where do we stand? In: Nentwig W, editor. Biological Invasions, Ecological

Studies Vol 193. Berlin: Springer-Verlag. pp. 97-125.

4. Ordonez A, Olff H (2013) Do alien plant species profit more from high resource

supply than natives? A trait-based analysis. Glob Ecol Biogeogr 22: 648-658.

5. Thompson K, Davis MA (2011) Why research on traits of invasive plants tells us

very little. Trends Ecol Evol 26: 155-156.

6. Nathan R, Müller-Landau HC (2000) Spatial patterns of seed dispersal, their

determinants and consequences for recruitment. Trends Ecol Evol 15: 278-285.

7. Sakai AK, Allendorf FW, Holt JS, Lodge DM, Molofsky J, et al. (2001) The

population biology of invasive species. Annu Rev Ecol Evol Syst 32: 305-332.

8. Pyšek P, Hulme PE (2005) Spatio-temporal dynamics of plant invasions: linking

pattern to process. Ecoscience 12: 302-315.

9. Lodge DM (1993) Biological Invasions: lessons for Ecology. Trends Ecol Evol 8:

133-137.

10. Davis MA, Grime JP, Thompson K (2000) Fluctuating resources in plant

communities: a general theory of invasibility. J Ecol 88: 528-534.

11. Davis MA (2003) Biotic globalization: does competition from introduced species

threaten biodiversity? Bioscience 53: 481-489.

12. Didham RK, Tylianakis JM, Hutchinson MA, Ewers RM, Gemmell NJ (2005) Are

invasive species the drivers of ecological change? Trends Ecol Evol 20: 468-

470.

13. Colautti RI, Grigorovich IA, MacIsaac HJ (2006) Propagule pressure: a null model

for biological invasions. Biol Invasions 8: 1023-1037.

42 Are introduced species better dispersers?

14. Pickering CM, Mount A (2010) Do tourists disperse weed seed? A global review of

unintentional human-mediated terrestrial seed dispersal on clothing, vehicles and

horses. J Sustainable Tour 18: 239-256.

15. Lake JC, Leishman MR (2004) Invasion success of exotic plants in natural

ecosystems: the role of disturbance, plant attributes and freedom from

herbivores. Biol Conserv 117: 215-226.

16. Cadotte MW, Lovett-Doust J (2001) Ecological and taxonomic differences between

native and introduced plants of southwestern Ontario. Ecoscience 8: 230-238.

17. Thompson K, Hodgson JG, Rich TCG (1995) Native and alien invasive plants: more

of the same? Ecography 18: 390-402.

18. Prinzing A, Durka W, Klotz S, Brandl R (2002) Which species become aliens? Evol

Ecol Res 4: 385-405.

19. Honig MA, Cowling RM, Richardson DM (1992) The invasive potential of

Australian banksias in South African fynbos: A comparison of the reproductive

potential of Banksia ericifolia and Leucadendron laureolum. Austral Ecol 17:

305-314.

20. Martin PH, Canham CD (2010) Dispersal and recruitment limitation in native versus

exotic tree species: life-history strategies and Janzen-Connell effects. Oikos 119:

807-824.

21. Säumel I, Kowarik I (2010) Urban rivers as dispersal corridors for primarily wind-

dispersed invasive tree species. Landsc Urban Plan 94: 244-249.

22. Daehler CC (2003) Performance comparisons of co-occurring native and alien

invasive plants: implications for conservation and restoration. Annu Rev Ecol

Evol Syst 34: 183-211.

43 Are introduced species better dispersers?

23. Macdougall AS, Turkington R (2005) Are invasive species the drivers or passengers

of change in degraded ecosystems? Ecology 86: 42-55.

24. Müller-Landau HC, Wright SJ, Calderón O, Condit R, Hubbell SP (2008)

Interspecific variation in primary seed dispersal in a tropical forest. J Ecol 96:

653-667.

25. Howe HF, Smallwood J (1982) Ecology of seed dispersal. Annu Rev Ecol Evol Syst

13: 201-228.

26. Moles AT, Warton DI, Warman L, Swenson NG, Laffan SW, et al. (2009) Global

patterns in plant height. J Ecol 97: 923-932.

27. Westoby M, Jurado E, Leishman M (1992) Comparative evolutionary ecology of

seed size. Trends Ecol Evol 7: 368-372.

28. Lord JM, Egan J, Clifford T, Jurado E, Leishman MR, et al. (1997) Larger seeds in

tropical floras: consistent patterns independent of growth form and dispersal

mode. J Biogeogr 24: 205-211.

29. Henery ML, Westoby M (2001) Seed mass and seed nutrient content as predictors

of seed output variation between species. Oikos 92: 479-490.

30. Aarssen LW, Jordan CY (2001) Between-species patterns of covariation in plant

size, seed size and fecundity in monocarpic herbs. Ecoscience 8: 471-477.

31. Westoby M, Falster DS, Moles AT, Vesk PA, Wright IJ (2002) Plant ecological

strategies: some leading dimensions of variation between species. Annu Rev

Ecol Evol Syst 33: 125-159.

32. Thomson FJ, Moles AT, Auld TD, Ramp D, Ren S, et al. (2010) Chasing the

unknown: predicting seed dispersal mechanisms from plant traits. J Ecol 98:

1310-1318.

44 Are introduced species better dispersers?

33. Thomson FJ, Moles AT, Auld TD, Kingsford RT (2011) Seed dispersal distance is

more strongly correlated with plant height than with seed mass. J Ecol 99: 1299-

1307.

34. Vilà M, Maron JL, Marco L (2005) Evidence for the enemy release hypothesis in

Hypericum perforatum. Oecologia 142: 474-479.

35. Ordonez A, Wright IJ, Olff H (2010) Functional differences between native and

alien species: a global-scale comparison. Funct Ecol 24: 1353-1361.

36. Daws M, Hall J, Flynn S, Pritchard HW (2007) Do invasive species have bigger

seeds? Evidence from intra- and inter-specific comparisons. S Afr J Bot 73: 138-

143.

37. Hawkes CV (2007) Are invaders moving targets? The generality and persistence of

advantages in size, reproduction, and enemy release in invasive plant species

with time since introduction. Am Nat 170: 832-843.

38. Thébaud C, Simberloff D (2001) Are plants really larger in their introduced ranges?

Am Nat 157: 231-236.

39. Maron JL, Vilà M, Bommarco R, Elmendorf S, Beardsley P (2004) Rapid evolution

of an invasive plant. Ecol Monogr 74: 261-280.

40. Mason RAB, Cooke J, Moles AT, Leishman MR (2008) Reproductive output of

invasive versus native plants. Glob Ecol Biogeogr 17: 633-640.

41. Tummers B (2006) DataThief III. Available: http://datathief.org/. Accessed: 2011

November.

42. Moles AT, Ackerly DD, Webb CO, Tweddle JC, Dickie JB, et al. (2005) Factors

that shape seed mass evolution. Proc Natl Acad Sci U S A 102: 10540-10544.

43. Royal Botanical Gardens Kew (2008) Seed Information Database (SID). Version

7.1. Available:http://data.kew.org/sid/ Accessed 2012 April.

45 Are introduced species better dispersers?

44. Richardson DM, Pyšek P, Rejmánek M, Barbour MG, Dane F, et al. (2000)

Naturalization and invasion of alien plants: concepts and definitions. Divers

Distrib 6: 93-107.

45. Vitousek PM, D'Antonio C, Loope LL, Rejmánek M, Westbrooks R (1997)

Introduced Species: a significant component of human-caused global change. N

Z J Ecol 21: 1-16.

46. Preacher KJ (2001) Calculation for the chi-square test: An interactive calculation

tool for chi-square tests of goodness of fit and independence. Available:

http://quantpsy.org. Accessed: 2013 March.

47. Pyšek P (1998) Is there a taxonomic pattern to plant invasions ? Oikos 82: 282-294.

48. Daehler CC (1998) The taxonomic distribution of invasive angiosperm plants:

ecological insights and comparison to agricultural weeds. Biol Conserv 84: 167-

180.

49. Jongejans E, Skarpaas O, Shea K (2008) Dispersal, demography and spatial

population models for conservation and control management. Perspect Plant

Ecol Evol Syst 9: 153-170.

50. Felsenstein J (1985) Phylogenies and the comparative method. Am Nat 125: 1-15.

51. Dostál P, Allan E, Dawson W, van Kleunen M, Bartish I, et al. (2013) Enemy

damage of exotic plant species is similar to that of natives and increases with

productivity. J Ecol 101: 388-399.

52. Chrobock T, Kempel A, Fischer M, van Kleunen M (2011) Introduction bias:

cultivated alien plant species germinate faster and more abundantly than native

species in Switzerland. Basic Appl Ecol 12: 244-250.

53. Williamson MH, Fitter A (1996) The varying success of invaders. Ecology 77:

1661-1666.

46 Are introduced species better dispersers?

54. Crawley MJ, Harvey PH, Purvis A (1996) Comparative ecology of the native and

alien floras of the British Isles. Philos Trans R Soc Lond B Biol Sci 351: 1251-

1259.

55. Thompson K, McCarthy Ma (2008) Traits of British alien and native urban plants. J

Ecol 96: 853-859.

56. Knapp S, Kühn I (2012) Origin matters: widely distributed native and non-native

species benefit from different functional traits. Ecol Lett 15: 696-703.

57. Küster EC, Durka W, Kühn I, Klotz S (2009) Differences in the trait compositions

of non-indigenous and native plants across Germany. Biol Invasions 12: 2001-

2012.

58. R Core Team (2013) R: a language and environment for statistical computing.

Available: http://www.r-project.org. Accessed: 2013 June.

59. Sokal RR, Rohlf FJ (1995) Biometry: the principles and practices of stastistics in

biological research. New York: W.H. Freeman. 887-887 p.

60. Bullock JM, Moy IL, Pywell RF, Coulson SJ, Nolan AM, et al. (2002) Plant

dispersal and colonization processes at local and landscape scales. In: Bullock

JM, Kenward RE, Hails RS, editors. Dispersal ecology: 42nd symposium of the

british ecological society. Oxford: Cambridge University Press. pp. 279-302.

61. Rejmánek M (1996) A theory of seed plant invasiveness: the first sketch. Biol

Conserv 78: 171-181.

62. Moravcová L, Pyšek P, Jarosik V, Havlícková V, Zákravský P (2010) Reproductive

characteristics of neophytes in the Czech Republic: traits of invasive and non-

invasive species. Preslia 82: 365-390.

63. Williamson MH, Fitter A (1996) The characters of successful invaders. Biol

Conserv 78: 163-170.

47 Are introduced species better dispersers?

64. Rejmanek M, Richardson DM (1996) What attributes make some plant species more

invasive? Ecology 78: 171-181.

65. Grotkopp E, Rejmánek M, Rost TL (2002) Toward a causal explanation of plant

invasiveness: seedling growth and life-history strategies of 29 Pine (Pinus)

species. Am Nat 159: 396-419.

66. Simberloff D (2009) The role of propagule pressure in biological invasions. Annu

Rev Ecol Evol Syst.

67. Trakhtenbrot A, Nathan R (2005) The importance of long‐distance dispersal in

biodiversity conservation. Diversity and ….

68. Kolar CS, Lodge DM (2001) Progress in invasion biology: predicting invaders.

Trends Ecol Evol 16: 199-204.

69. van Kleunen M, Johnson SD (2007) South African Iridaceae with rapid and profuse

seedling emergence are more likely to become naturalized in other regions. J

Ecol 95: 674-681.

70. van Kleunen M, Dawson W, Schlaepfer D, Jeschke JM, Fischer M (2010) Are

invaders different ? A conceptual framework of comparative approaches for

assessing determinants of invasiveness. Ecol Lett 13: 947-958.

71. Richardson DM, Pyšek P (2012) Naturalization of introduced plants: ecological

drivers of biogeographical patterns. New Phytol 196: 383-396.

Supporting information

48 Are introduced species better dispersers?

Figure S1. Graphs of relationships between dispersal distance of introduced vs. native species when accounting for seed mass, plant height or dispersal syndrome individually.

Table S1. Attributes of introduced and native species.

Table S2. Details of analyses of dispersal distance of native vs. introduced species when accounting for seed mass, plant height or dispersal syndrome, individually.

Table S3. Comparisons of introduced and native species‟ seed dispersal distances using a subset of data with no missing values.

Table S4. Total number of missing values for trait data.

Supporting information S1. Analyses including a random effect for site.

49 Are introduced species better dispersers?

Figure S1: Graphs of relationships between dispersal distance of introduced vs. native species when accounting for seed mass, plant height or dispersal syndrome individually.

1) Black circles and dashed lines represent introduced species, while white circles and complete lines represent native species. Ps represents the P values of the species status effect on the model. Pint represents the P value of the interaction of seed mass and species status (panels A and B) or plant height and species’ status (panels C and D) effect on the model.

A C 10000 10000 1000 1000 100 100 10 10

[log scale] [log 1 1

0.1 P = 0.43 0.1 P = 0.3 s s

Mean dispersal distance (m) distance dispersal Mean 0.01 P = 0.37 0.01 P = 0.03 int int 1

10 0.1 1 10 100 0.1 100 0.01 1000 0.001 10000 0.0001 100000 1000000 10000000

10000 B 10000 D 1000 1000 100 100 10 10

[log scale] [log 1 1 0.1 P = 0.80 0.1 P = 0.86 s s 0.01 P = 0.05 0.01 P = 0.001 int int Maximum(m) distance dispersal 1

10 0.1 1 10 100 0.1 100 0.01 1000

0.001 Plant height (m) [log scale] 10000 0.0001 100000 1000000 10000000 Seed mass (mg) [log scale]

50 Are introduced species better dispersers?

2) Differences between native and introduced species’ seed dispersal distances for species with animal dispersal, wind/water dispersal or unassisted dispersal. Black dashed lines represent mean values. The boxes represent 25th, 50th and 75th percentiles.

Whiskers represent the 10th and 90th percentiles, outliers are shown as points. Letters above boxes represent dispersal syndromes with significantly (P < 0.05) different dispersal distances. Numbers below boxes are sample sizes. Ps represents the P values of the species status effect on the model. Pint represents the P value of the interaction of dispersal syndrome and species’ status on the model.

P A s = 0.48 P 10000 int = 0.10 A B C 1000

100

10

[log scale] 1

0.1

Mean dispersal distance (m) 0.01 7 196 26 52 15 55 Introduced Native Introduced Native Introduced Native Animal Water/wind Unassisted

P B s = 0.008 P 100000 int = 0.003 10000 A B C

1000

100

10 [log scale] 1

0.1

0.01 Maximum dispersal distance (m) 10 143 26 60 17 51 Introduced Native Introduced Native Introduced Native Animal Wind Unassisted

51 Are introduced species better dispersers?

3) Residuals vs. fitted values plots for the mean dispersal distance comparison of introduced and native species accounting for plant height (A) and seed mass (B), and residuals vs. fitted values plots for the maximum dispersal distance comparison of introduced and native species accounting for plant height (C) and seed mass (D). Note the lack of a well-defined pattern in the four plots suggesting homogeneity of variance, and thus the suitability of these data for ANCOVA analysis.

2 A B 2 1 1

0 0 Residuals -1 -1

-2 -2

-1 0 1 2 0.0 0.5 1.0 1.5 2.0

C D 2 2

1 1

0 0 Residuals

-1 -1

-2 0.5 1.0 1.5 2.0 2.5 3.0 1.0 1.2 1.4 1.6 Fitted values Fitted values

52 Are introduced species better dispersers?

Table S1. Attributes of introduced and native species.

Observed number (and percent) of introduced and native species in each category for dispersal syndrome, taxonomic group, nitrogen fixation, growth form and biome. A species could be counted in more than one category for dispersal syndrome, but not for any of the other factors. Taxonomy follows the Angiosperm Phylogeny Website [1].

Proteales, Vitales and Gymnosperms (each represented by few species) were grouped to avoid having cells with expected values less than 1, which invalidates the Chi-square test [2].

Species’ status

Dispersal syndrome Native Introduced

Animal 232 (62.2%) 10 (17.2%)

Unassisted 60 (16.1%) 17 (29.3%)

Water/Wind 81 (21.7%) 31 (53.4%)

Taxonomic group Native Introduced (56)

Monocots 31 (8.6%) 8 (14.3%)

Magnoliids 34 (9.4%) 1 (1.8%)

Ranunculales 6 (1.7%) 2 (3.6%)

Core 21 (5.8%) 4 (7.1%)

(Caryophyllales,

Saxifragales, Santalales)

Rosids I (Fabidae) 116 (32.0%) 8 (14.3%)

Rosids II (Malvidae) 52 (14.4%) 10 (17.8%)

53 Are introduced species better dispersers?

Basal (Ericales) 17 (4.6%) 2 (3.6%)

Asterids I (incl Lamiales, 39 (10.8%) 8 (14.3%)

Gentianales, Solanales)

Asterids II (incl Asterales, 33 (9.1%) 12 (21.4%)

Apiales)

Other (Proteales, Vitales, 13 (3.6%) 1 (1.8%)

Gymnosperms)

Nitrogen fixation Native (362) Introduced (56)

Present 27 (7.5%) 5 (9.8%)

Absent 333 (92.5%) 46 (90.2%)

Growth form Native Introduced

Woody 227 (63.1%) 16 (31.4%)

Non-woody 133 (36.9%) 35 (68.6%)

Biome Native Introduced

Grassland 81 (22.5%) 23 (45.1%)

Shrubland 19 (5.3%) 1 (2%)

Temperate forest 58 (16.1%) 7 (13.7%)

Tropical forest 146 (40.6%) 4 (7.8 %)

Woodland 42 (11.7%) 6 (11.8%)

Other/unknown 14 (3.9%) 10 (19.6%)

54 Are introduced species better dispersers?

References

1. Stevens PF (2001 onwards) Angiosperm Phylogeny Website.

http://www.mobot.org/MOBOT/research/APweb/.

2. Preacher KJ (2001) Calculation for the chi-square test: An interactive calculation tool

for chi-square tests of goodness of fit and independence. Available:

http://quantpsy.org. Accessed: 2013 March.

55 Are introduced species better dispersers?

Table S2. Details of analyses of dispersal distance of native vs. introduced species when accounting for seed mass, plant height or dispersal syndrome, individually.

1) MEAN SEED DISPERSAL DISTANCE AND SEED MASS

Term Estimate Standard error Sum of Squares d.f. P

Intercept 0.3 0.09 10 14 0.002

Seed mass 0.33 0.05 47.88 1 < 0.0001

Status 0.18 0.23 0.65 1 0.43

Status × seed mass 0.14 0.16 0.84 1 0.37

Residuals 243.25 233

2) MAXIMUM SEED DISPERSAL DISTANCE AND SEED MASS

Term Estimate Standard error Sum of Squares d.f. P

Intercept 1.17 0.08 180 1 < 0.0001

Seed mass 0.13 0.04 7.1 1 0.005

Status 0.05 0.16 0.06 1 0.80

Status × seed mass 0.28 0.1 3.58 1 0.05

Residuals 216.46 245

56 Are introduced species better dispersers?

3) MEAN SEED DISPERSAL DISTANCE AND PLANT HEIGHT

Term Estimate Standard error Sum of Squares d.f. P

Intercept 0.24 0.06 8.12 1 < 0.0001

Plant height 0.97 0.06 140.85 1 < 0.0001

Status 0.15 0.15 0.57 1 0.3

Status × plant height 0.47 0.21 2.58 1 0.03

Residuals 116.19 228

4) MAXIMUM SEED DISPERSAL DISTANCE AND PLANT HEIGHT

Term Estimate Standard Sum of d.f. P

error Squares

Intercept 1.09 0.06 195.81 1 < 0.0001

Plant height 0.46 0.07 28.07 1 < 0.0001

Status 0.03 0.15 0.02 1 0.86

Status × plant height 0.83 0.21 9.24 1 0.0001

Residuals 146.08 236

5) MEAN SEED DISPERSAL DISTANCE AND DISPERSAL SYNDROME

Term Sum of Squares d.f. P

Intercept 344 1 < 0.0001

Dispersal syndrome 104.66 2 < 0.0001

Status 2.28 1 0.48

Dispersal syndrome × status 1.53 2 0.1

Residuals 291.68 290

57 Are introduced species better dispersers?

6) MAXIMUM SEED DISPERSAL DISTANCE AND DISPERSAL SYNDROME

Term Sum of Squares d.f. P

Intercept 397.03 1 < 0.0001

Dispersal syndrome 50.9 2 < 0.0001

Status 4.55 1 0.008

Dispersal syndrome × status 7.75 2 0.003

Residuals 195.55 301

7) ESTIMATES FOR DISPERSAL SYNDROME ANALYSES

Mean dispersal distance Maximum dispersal distance

Term Estimate Standard error Estimate Standard error

Intercept 1.51 0.07 1.6663 0.0674

Unassisted -1.7 0.15 -1.1461 0.1315

Water/wind -1.24 0.16 -0.4837 0.124

Introduced 0.27 0.39 0.6978 0.2636

Unassisted × status -0.42 0.49 -1.1682 0.3471

Water/wind × status 0.41 0.46 -0.4753 0.3245

58 Are introduced species better dispersers?

Table S3. Comparisons of introduced and native species’ seed dispersal distances using a subset of data with no missing values.

Status refers to whether a species is native or introduced. All analyses were done on log-transformed data.

1) ANALYSES WITH NO ADDITIONAL PREDICTORS

Mean dispersal distance Maximum dispersal distance

Term t d.f. P t d.f. P

Status -0.42 41.2 0.7 -0.68 47.1 0.5

2) ANALYSES WITH ONE ADDITIONAL PREDICTOR

Mean dispersal distance Maximum dispersal distance

Term Sum of d.f. P Sum of d.f. P

squares squares

Status 0.09 1 0.7 0.25 1 0.6

Seed mass 31.6 1 <0.0001 4.34 1 0.02

Status × seed mass 2.6 1 0.09 2.8 1 0.06

Residuals 183.9 208 172.11 223

Status 0.5 1 0.3 0.000 1 1

Plant height 101.6 1 <0.0001 24.8 1 <0.0001

Status × plant height 2.7 1 0.02 5.4 1 0.004

59 Are introduced species better dispersers?

Residuals 99.8 208 139.3 223

Status 3.06 1 0.05 4.8 1 0.003

Dispersal syndrome 52.8 2 <0.0001 33.3 2 <0.0001

Status × dispersal syndrome 2.9 2 0.2 4.5 2 0.02

Residuals 157.9 206 114.8 215

3) FULL MODEL

Mean dispersal distance Maximum dispersal distance

Term Sum of d.f. P Sum of d.f. P

squares squares

Species status 0.08 1 0.7 0.6 1 0.2

Seed mass 2.9 1 0.01 1.5 1 0.06

Plant height 30.7 1 <0.0001 3.4 1 <0.01

Dispersal syndrome 5.3 2 <0.01 23.1 2 <0.0001

Species status × Dispersal 0.15 2 0.8 0.25 2 0.7 syndrome

Species status × Seed mass 0.3 1 0.4 4.4 1 <0.01

Species status × Plant height 0.12 1 0.6 3.6 1 <0.01

Species status × Plant height 2.7 4 0.2 8.5 4 <0.0001

× Dispersal syndrome

Species status × Seed mass × 2.1 4 0.3 8.1 4 <0.0001

Dispersal syndrome

Residuals 83.4 194 82.8 203

60 Are introduced species better dispersers?

Table S4. Present and missing values for trait data.

Number (and percent) of introduced and native species’ present/missing data for plant height, seed mass and dispersal syndromes. A species could be counted in more than one category for dispersal syndrome, but not for any of the other factors.

1) Present/missing data for plant height data.

Species’ status

Present/missing data home introduced

Missing 93 (25.8%) 11 (21.6%)

Present 267 (74.2) 40 (78.4%)

2) Present/missing data for seed mass data.

Species’ status

Present/missing data home introduced

Missing 96 (26.7%) 2 (3.9%)

Present 264 (73.3%) 49 (96.1%)

3) Present/missing data for dispersal syndrome data.

Species’ status

Present/missing data Home introduced

Missing 0 (0%) 0 (0%)

animal 232 (62.2%) 10 (17.2%)

unassisted 60 (16.1%) 17 (29.3%)

Water/wind 81 (21.1%) 31 (53.4)

61 Are introduced species better dispersers?

Supporting Information S1. Analyses including a random effect for site.

To test the influence of site on the dispersal distance comparison between introduced and native species we ran a linear mixed model were our predictor variables were plant height, seed mass, dispersal syndrome and a random term for site and our dependent variables were mean and maximum dispersal distance. Analyses were done on log- transformed dispersal distance, seed mass and plant height data. Plant height explained

1.2 times more variation than did dispersal syndrome, and almost twice as much variation as did seed mass (Table S1.1). Species’ status explained approximately five times less variation in mean dispersal distance than did plant height (Table S1.1). The term that explained the most variation in maximum dispersal distance was dispersal syndrome, followed by plant height, seed mass and the interaction between dispersal syndrome and species’ status (Table S1.2). Species’ status explained eleven times less variation in maximum dispersal distance than did dispersal syndrome (Table S1.2).

62 Are introduced species better dispersers?

Table S1.1. Comparison between introduced and native species’ mean dispersal distance including a random effect for site.

Source Estimate P

Species status 0.16 0.99

Seed mass -0.05 0.36

Plant height 0.63 < 0.0001

Unassisted dispersal syndrome -0.41 0.02

Wind/water dispersal syndrome -0.11 0.44

Species status × Unassisted dispersal syndrome -0.18 0.99

Species status × Wind/water dispersal syndrome -0.18 0.99

Species status × Seed mass -3.14 0.87

Species status × Plant height 7.33 0.85

Native status × Seed mass × Unassisted dispersal syndrome 0.09 0.42

Native status × Seed mass × Wind/water dispersal syndrome -0.33 0.02

Introduced status × Seed mass × Unassisted dispersal syndrome 3.22 0.87

Introduced status × Seed mass × Wind/water dispersal syndrome 3.54 0.86

Native status × Plant height × Unassisted dispersal syndrome -0.17 0.43

Native status × Plant height × Wind/water dispersal syndrome 0.79 <0.001

Introduced status × Plant height × Unassisted dispersal syndrome -7.16 0.86

Introduced status × Plant height × Wind/water dispersal syndrome -6.37 0.87

63 Are introduced species better dispersers?

Table S1.2 Comparison between introduced and native species’ maximum dispersal distance including a random effect for site.

Source Estimate P

Species status -0.72 0.75

Seed mass -0.01 0.62

Plant height 0.53 <0.0001

Unassisted dispersal syndrome -0.54 0.001

Wind dispersal syndrome -0.05 0.73

Species status × Unassisted dispersal syndrome 0.60 0.76

Species status × Wind dispersal syndrome 0.63 0.77

Species status × Seed mass -10.66 0.55

Species status × Plant height 22.35 0.54

Native status × Seed mass × Unassisted dispersal syndrome 0.06 0.40

Native status × Seed mass × Wind dispersal syndrome -0.48 <0.001

Introduced status × Seed mass × Unassisted dispersal syndrome 10.85 0.54

Introduced status × Seed mass × Wind dispersal syndrome 11 0.54

Native status × Plant height × Unassisted dispersal syndrome -0.08 0.56

Native status × Plant height × Wind dispersal syndrome 0.62 0.005

Introduced status × Plant height × Unassisted dispersal syndrome -22.10 0.54

Introduced status × Plant height × Wind/water dispersal syndrome -21.37 0.55

64 Recruitment of introduced and native species

Chapter two

A comparison of the recruitment success of introduced and

native species under natural conditions.

Habacuc Flores-Moreno & Angela T. Moles

PLoS ONE (2013)

The study was conceived and designed by HFM and ATM. HFM performed the

experiments and collated the data. HFM and ATM analysed the data. The manuscript

writing was led by HFM, with a substantial contribution by ATM.

This chapter has been modified from the original paper specifically for this thesis.

65 Recruitment of introduced and native species

Abstract

It is commonly accepted that introduced species have recruitment advantages over native species. However, this idea has not been widely tested, and those studies that have compared survival of introduced and native species have produced mixed results. We compiled data from the literature on survival through germination (seed to seedling survival), early seedling survival (survival through one week from seedling emergence) and survival to adulthood (survival from germination to first reproduction) under natural conditions for 285 native and 63 introduced species. Contrary to expectations, we found that introduced and native species do not significantly differ in survival through germination, early seedling survival, or survival from germination to first reproduction.

These comparisons remained non-significant after accounting for seed mass, longevity and when including a random effect for site. Results remained consistent after excluding naturalized species from the introduced species data set, after performing phylogenetic independent contrasts, and after accounting for the effect of life form (woody/non-woody).

Although introduced species sometimes do have advantages over native species (for example, through enemy release, or greater phenotypic plasticity), our findings suggest that the overall advantage conferred by these factors is either counterbalanced by advantages of native species (such as superior adaptation to local conditions) or is simply too small to be detected at a broad scale.

66 Recruitment of introduced and native species

Introduction

Seeds and seedlings are exposed to many risks during establishment, such as predation, loss of viability in the soil, drought, herbivory, pathogen attack, shading, nutrient deprivation and competition [1]. As a result, seeds and seedlings experience the highest mortality rates of any life history stage [2]. A widely accepted hypothesis in invasion ecology is that introduced or invasive species have higher survival through the early stages of establishment than do native or non-invasive species [3-5]. Introduced species are often associated with dominance, rapid spread and fast population growth rates [6], and it is feasible that higher survival of introduced species through the early stages of establishment could contribute to these.

The idea that introduced species might have higher recruitment success is based on both theoretical arguments and empirical observations of three main mechanisms: enemy release, higher plasticity, and faster growth rates. Enemy release could allow introduced species to escape from specialist seed predators and herbivores in their novel range [7].

However, evidence for enemy release has been inconsistent [8-10]. Greater plasticity of introduced species could allow them to germinate under a wider range of environmental conditions and promote their higher tolerance to environmental stress in the early stages of life. However, evidence for introduced species’ greater plasticity shows conflicting results

[11-14]. Faster growth rates could shorten the time introduced species spend in early vulnerable stages of the life cycle and/or reduce the time to reproduction, thus reducing a species’ exposure to mortality [4,15]. However, the available data for this is also varied

67 Recruitment of introduced and native species

[11,16-18]. That is, although some theories and data suggest that introduced species should have advantages over native species, the evidence has been inconsistent. Moreover, the magnitude of the positive effect of these mechanisms on introduced species’ fitness, specifically in the form of survival through the different stages of recruitment, is unclear.

Our main objective is to provide a large scale test of the idea that introduced species have superior recruitment success than do native species.

It has been suggested that introduced species germinate faster, to a higher percentage and in a wider range of conditions than do native species [5] and that high germination success promotes invasiveness [3]. Empirical studies have found higher germination percentages in introduced or invasive species [19-24], lower germination in introduced or invasive species

[25-27], mixed results [28-31] and non-significant differences in germination between introduced or invasive and native or non-invasive species [32-34]. These studies used a wide variety of approaches, but broadly can be divided into studies performed under natural conditions without manipulation, and studies performed under experimental or artificial conditions (e.g. in greenhouses, with supplementary watering, or herbivore exclosures).

The latter have described the proportion of germination across an amazing range of species under experimental or artificial conditions (e.g. [20,21,34]). Conversely, studies performed under natural conditions without manipulation have been commonly limited to few subject species (e.g. [19,23,31]). Thus, many studies have compared germination of introduced and native species; however nobody has ever compared the survival through germination of introduced and native species under natural conditions on a broad scale. This is our first aim.

68 Recruitment of introduced and native species

Pyšek and Richardson [3] proposed that higher rates of seedling survival and establishment should promote invasiveness in introduced species. However these authors highlight the lack of large comparative datasets available to test this idea. Some evidence shows that introduced or invasive species have higher seedling survival than natives [35,36]. However, several studies have found non-significant differences in seedling survival between native and introduced or invasive species [19,37-40]. Others have found that native species have higher seedling survival than do invasive species [23,41], while some have found mixed results [28,42,43]. One possible explanation for the abundance of mixed results is that the majority of these studies focus on small numbers of species, usually from one region.

Furthermore, many studies compare seedling survival and/or establishment of introduced and native species in greenhouse, laboratory conditions, or garden experiments where herbivory, water/light-stress and competition are often controlled for. Although studies under these conditions tell us what introduced species are capable of doing compared to native species, everything else being equal, they do not tell us about how introduced species are dealing with natural environmental conditions, selective pressures and interactions.

Thus, our second aim is to compare the early seedling survival of introduced and native species at a broad scale under natural conditions.

Seed mass is an ecologically important trait that affects the recruitment strategy of plants

[1]. Plants with bigger seeds have higher early seedling survival than do those with smaller seeds [44]. Thus, it is possible that a difference in seed mass between native and introduced

69 Recruitment of introduced and native species

species could mask or generate differences in early seedling survival between introduced and native species. Therefore our third aim is to assess whether introduced and native species differ in early seedling survival under natural conditions, once the effect of seed mass had been accounted for. No significant relationship has been found between seed mass and proportion of germination, or between seed mass and survival from seedling to reproduction [44], and the advantages of large seed size are known to be restricted to early establishment –usually no later than cotyledon phase [45]. Therefore, we did not control for seed mass in our assessment of differences in germination or survival from seed to fruiting.

If introduced species do not survive until reproduction, then no naturalized or introduced population can succeed. Previous studies comparing survival from germination to first reproduction between introduced and native species have produced contrasting results, including, higher survival in introduced or invasive species[46], mixed results [47,48] and lower survival of invasive species’ in their introduced range [46,47,49,50]. However, all previous work has been on pairs or small groups of species. The fourth aim of our study is to ask whether introduced species have higher survival from germination to first reproduction than do native species at a broad scale, under natural conditions.

It has been proposed that short life cycles will be favoured among introduced species

[4,51]. If this is the case then differences in longevity between introduced and native species could overshadow differences in recruitment by affecting the time over which species are exposed to mortality. Our fifth aim is to assess whether introduced and native

70 Recruitment of introduced and native species

species differ in survival through germination, early seedling survival, and survival from germination to first reproduction once the effect of longevity has been taken into account.

For the comparison of early seedling survival between introduced and native species we first accounted for the effect of longevity by itself; then we ran a model that also included a term for seed mass.

In summary, the hypotheses that we addressed in this paper were:

1) Introduced species have higher survival through germination (seed to

seedling survival) than do native species.

2) Introduced species have a higher early seedling survival (survival for the

first week after germination) than do native species.

3) Introduced species have higher early seedling survival than do native species

once seed mass has been accounted for.

4) Introduced species have higher survival from germination to first

reproduction than do native species.

5) Introduced species have higher recruitment success than do native species

once the effect of longevity has been accounted for.

71 Recruitment of introduced and native species

Methods

Ethical statement

No permission or approval was required for obtaining the data included in this study because all the data were extracted from published sources.

We began with the database for germination, seedling survival and seedling survival from germination to first reproduction generated by Moles et al. [44]. This database contains studies published up to week 38, 2002 with information on germination and seedling survival. Therefore, we searched ISI web of knowledge for papers in English from week

38, 2002 to week 3, 2012 containing the words ‘seedling survival’ and ‘germination’ restricted by the terms ‘weed$’, ‘introduced’, ‘invasive’, ‘non-invasive’, ‘naturali*’, ‘alien’,

‘non-native’, and ‘noxious’ in order to obtain more data points for introduced species’ germination, seedling survival, and survival from germination to first reproduction.

Only studies measuring the survival of seeds and seedlings of both native and introduced species in natural conditions were included. Studies were excluded if seedling were raised in pots, within exclosures, under shelters, with extra watering, with pesticides, with weeding, or with supplementary fertilization. Studies were also excluded if individual seedling survival was not followed from the day of emergence or if they were transplanted after emergence. Only studies with a minimum sample size of ten individuals were included. In total our database contained information for 348 species from 90 families,

72 Recruitment of introduced and native species

from 186 different sites around the world. This included data for 285 native species and 63 introduced species. Of the 63 introduced species in our database, two were classified as naturalized and 61 as invasive.

We extracted information on survival through germination (seed to seedling survival), early seedling survival (survival through one week from seedling emergence), survival from germination to first reproduction, seed mass and introduced/native status. These data were extracted from three sources from the papers in our database, with the following order of preference: 1) tables, 2) text and 3) graphs (using Datathief III [52]). Additional data on introduced/native status were compiled from the global compendium of weeds [53] and environmental agencies from the regions where the studies took place. Additional seed mass data were compiled from Moles et al. [54] and Kew botanical gardens’ seed information database [55].

Ideally, longevity should be measured as a continuous variable. Unfortunately, continuous longevity data are very scarce. We collected categorical longevity data (annual, biennial and perennial; lifespan categories) for 345 species, and continuous maximum recorded longevity data (continuous longevity) for 128 species. Lifespan categories data were collected in decreasing order of preference from 1) the same papers as survival data, 2) environmental agencies from the region in which the study took place or 3) papers found in main references. Continuous longevity data were collected from the global literature (Table

S6). We have lifespan categories data for 99.7 % of our species, and continuous longevity

73 Recruitment of introduced and native species

data for 37% of our species. That is, the strength of the lifespan categories data is coverage, while the strength of the continuous longevity data is resolution.

Before analysis we log10- transformed the seed mass and continuous longevity data, and logit-transformed data for survival through germination, early seedling survival, and survival from germination to first reproduction [56]. To avoid problems with species with survival values equal to 0 or 1 (0 or 100% survival respectively) we added or subtracted the smallest non-zero value within each of the recruitment stages to these species [56]. Species recorded both in a native and introduced region were statistically weighted such that each species had a total statistical weight of one.

To determine whether introduced species have higher survival through germination, early seedling survival (survival for the first week after germination) and survival from germination to first reproduction than do native species, we ran t-tests assuming unequal variance with species’ status (introduced/native) as our predictor variable and survival through germination, early seedling survival, or survival from germination to first reproduction as our dependent variables.

To determine whether introduced species have higher early seedling survival than do native species once the effect of seed mass has been accounted for, we used a linear model where

74 Recruitment of introduced and native species

the predictor variables were species’ status (introduced/native) and seed mass, and the dependent variable was early seedling survival.

To determine whether introduced and native species differ in survival through germination, early seedling survival, and survival from germination to first reproduction after accounting for the effect of longevity we ran a linear model where our predictor variables were species’ status (introduced/native) and lifespan categories (annual, biennial and perennial life cycles), and our dependent variables were survival through germination, early seedling survival and seedling survival to first reproduction. For early seedling survival, we also ran a linear model that accounted for the effect of species’ status, seed mass, lifespan categories and their interactions. In this linear model the predictor variables were species’ status (introduced/native), seed mass and lifespan categories (annual, biennial and perennial life cycles) and the dependent variable was early seedling survival. The low number of biennial species with data for early seedling survival (n = 7), and survival from germination to first reproduction (n = 5) affected the number of degrees of freedom needed for these linear models. In our study annual, biennial and perennial plants did not significantly differ in early seedling survival or survival from germination to first reproduction. In order to have a statistically more powerful comparison we merged the biennial and perennial species into one lifespan category. These results do not qualitatively differ from comparisons where biennial plants have been excluded (Table S1). Finally, we compared the continuous longevity of introduced and native species using a t-test and re-ran all longevity analyses using continuous longevity data (Table S6). All analyses were performed in R 2.15.1 [57].

75 Recruitment of introduced and native species

Data considerations

In our main analyses, we asked whether differences in seed mass and longevity between native and introduced species might have affected our results. Here, we consider some additional factors: site to site variation, degree of invasiveness of the study species, phylogeny, and life form (woody/non-woody).

1. Site to site variation

Some data points in our database come from the same site. To explicitly account for their non-independence we ran all analyses using mixed models including terms for: lifespan categories (annual, biennial and perennial), status (introduced/native), and a random term for site. The model for early seedling survival also included a term for seed mass because early seedling survival is affected by it ([44]; see above).

2. Naturalized and invasive introduced species

Not all introduced species are invasive. However of the 63 introduced species included in our study, two were classified as naturalised and 61 were classified as invasive (sensu [58]) by the primary studies or environmental agencies of the countries were the studies took place. Investigating the effect of species’ invasiveness on recruitment was beyond the scope of our study. However, we did ask whether our inclusion of non-invasive introduced species might have obscured a significant result.

76 Recruitment of introduced and native species

3. Phylogeny

Phylogenetic analyses investigate the evolution of traits. These analyses explicitly assume that the traits under comparison are heritable [59]. Species’ status (introduced/native) and survival are non-heritable ecological traits. Species do not evolve to be introduced (i.e. the change from native to introduced state does not evolve along a phylogeny) and while some traits associated with survival are heritable, survival also depends on non-heritable factors such as environmental conditions, intensity of interactions with herbivores, predators, pathogens and other plants, and chance. Thus, phylogenetic analyses are not appropriate for our data. Nevertheless, to dispel any doubts about the robustness of our results stemming from the potential non-independence of data points due to phylogenetic relatedness, we performed a phylogenetic independent contrast analysis (Table S3). First, we constructed a phylogeny of plant species included in our dataset using PHYLOMATIC ([60,61];

PHYLOMATIC tree version R20100701), and generated phylogenetically independent contrasts using the Analysis of Traits module in PHYLOCOM 4.2 [62]. Finally, we used one-sample t-tests to determine whether changes in species’ status had been consistently associated with changes in survival rate through the evolutionary history of these species.

4. Life-form (woody/non-woody)

We tested whether there was a difference in proportion of introduced and native species with woody or non-woody growth form using a Chi squared test. Then, we ran linear models where our predictor variables were species status (introduced/native) and life-form

77 Recruitment of introduced and native species

(woody/non-woody) and our dependent variables were survival through germination, early seedling survival, and survival from germination to first reproduction.

78 Recruitment of introduced and native species

Results

Contrary to expectations, we did not find differences in introduced and native species’ survival through germination (seed to seedling survival; P = 0.36; Fig. 1A), early seedling survival (one week survival from seedling emergence; P = 0.85; Fig. 1B), or survival from germination to first reproduction (P = 0.22; Fig. 1C). On average, 12% of the seeds that enter germination survive to seedling, 94% of the individuals that enter the seedling stage survive for one week and only 8% of seedlings survive from germination to first reproduction.

After controlling for lifespan categories, the differences between introduced and native species’ survival through germination (P = 0.34; Table S2), early seedling survival (P =

0.84; Table S2) and survival from germination to first reproduction (P = 0.75; Table S2) remained non-significant. Lifespan categories explained substantially more variation in survival through germination (2.7-fold), early seedling survival (3-fold), and survival from germination to first reproduction (9-fold) than did species’ status. When we compared the continuous longevity of introduced and native species we found that, on average, introduced species had significantly shorter lifespans (~9.6 years, P = 0.0005) than did native species (~53.6 years; Table S6). However, we still did not find a significant difference on survival through germination (P = 0.07), early seedling survival (P = 0.26), or survival from germination to first reproduction (P = 0.65) between introduced and native species (Table S6).

79 Recruitment of introduced and native species

80 Recruitment of introduced and native species

Fig. 1. Comparison of introduced and native species recruitment success. Introduced and native species’ (A) survival through germination, (B) early seedling survival (one week survival after germination) and (C) survival from germination to first reproduction. Black dashed lines represent geometric mean values. Boxes represent the 25th, 50th and 75th percentiles. Whiskers represent the 10th and 90th percentiles, outliers are represented as points.

We did not find a significant difference between native and introduced species’ early seedling survival after controlling for seed mass (P = 0.49; Table S1). When we compared the early seedling survival of native and introduced species once the effect of lifespan categories, seed mass and their interactions had been accounted for, we did not find a significant effect of species’ status on early seedling survival (P = 0.27; Table S2).

However, we found a significant positive effect of lifespan categories (P = 0.002, Table

S2). Lifespan categories explained far more variation (R2 = 0.07) than did either species’ status (R2 = 0.008) or the interaction between species’ status and lifespan categories (R2 =

0.01). Results remained qualitatively similar when using continuous longevity instead of lifespan categories (Table S6).

Data considerations

1. Site to site variation

81 Recruitment of introduced and native species

Analyses including a random effect for site were broadly consistent with previous results, showing that species’ status does not have a significant effect on survival through germination (P = 0.18; Table S3), early seedling survival (P = 0.48; Table S3), or survival from germination to first reproduction (P = 0.79; Table S3). Overall, site accounted for

31%, 40% and 6% of the variation in survival through germination, early seedling survival and seedling survival to reproduction respectively. This is in line with the results of previous broad scale data compilations, where site variation has typically explained about half of the observed variation (e.g. [63-65]).

2. Naturalized and invasive introduced species

After excluding the two naturalized species from our database the difference between introduced and native species’ survival through germination, early seedling survival, and survival from germination to first reproduction remained non-significant (P = 0.28, 0.59 and 0.22, respectively).

3. Phylogeny

Like our cross-species analyses, phylogenetic analyses showed no significant difference between introduced and native species’ survival through germination (P = 0.91; Table S4), early seedling survival (P = 0.32; Table S4), or survival from germination to first reproduction (P = 0.65; Table S4).

82 Recruitment of introduced and native species

4. Life-form (woody/non-woody)

We found that native species are significantly more likely to be woody (Χ2 = 12.92, d.f. = 1,

P = 0.0003; Table S5). However, we found no significant effect of species’ status on survival through germination (P = 0.47; Table S5), early seedling survival (P = 0.72; Table

S5), and survival from germination to first reproduction (P = 0.48; Table S5) once the effect of life-form (woody/non-woody) was accounted for. Only in the case of early seedling survival was there a significant effect of life form (P = 0.02; Table S5).

Discussion

Our most important finding is that introduced and native species do not significantly differ in survival through germination, early seedling survival, and survival from germination to first reproduction. These results remained non-significant after accounting for seed mass, longevity, when including a random effect for site, excluding non-invasive introduced species, accounting for phylogeny, or accounting for life form. These findings were contrary to our expectations, and also to expectations in the literature [3,5,20,21,34]. Our results suggest that the idea that high germination success, seedling survival, and establishment are key drivers of introduced species’ spread and dominance in new environments is not universal across introduced species and needs to be carefully assessed for individual species. After all, given that the native species have had generations of selection for traits that favour survival under the local conditions we may have expected native species to exceed, or at least match, the survival of introduced species. However, there would be great value in determining whether recruitment success differs between

83 Recruitment of introduced and native species

introduced and native species under experimental settings. This could be done using a meta-analytical approach, because experimental data normally consists of at least one control treatment. The control treatments could be compared across studies allowing the correct calculation of a effect size index or response ratio [66,67].

In this paper we compared introduced and native species survival through germination, early seedling survival and from germination to first reproduction because it is through these early stages that plants experience the strongest selective pressures in the form of high rates of mortality [1,2]. However, there is whole range of processes occurring after recruitment by which introduced species could benefit from. For instance, several studies have proposed that introduced species could benefit from enemy release, faster growth rate or phenotypic plasticity. Our data suggest that any advantage introduced species accrue as a result of these factors is either small enough in magnitude or uncommon enough that it does not make a detectable difference to overall survival. One possibility is that the advantages of introduced species are balanced by the superior adaptation of natives to their environment. Our results are consistent with recent meta-analyses that have found that enemy release [8,9] and plasticity in invasive and/or introduced species do not always result in higher fitness or performance [12]. In the case of plasticity, our result could also be consistent with introduced species’ not having higher plasticity in the first place (see [13], for a more detailed discussion see [14,68]). Evidence for higher relative growth rates is mixed [11,16-18], and no clear relationship between relative growth rate and survival to adulthood has yet been described.

84 Recruitment of introduced and native species

Many studies have shown significant differences in functional or morphological traits of introduced and native species (e.g. [3,4,11,16,18]). These traits are often chosen for study because they are thought to be indicators of important processes such as seed production, seed dispersal, recruitment and growth. However, our study shows no link between species’ status (introduced or native) and higher success through recruitment. This discrepancy between the findings of trait-based studies (e.g. [16,18]) and process-based studies (e.g.

[69]) merits further investigation. That is, invasion ecology needs to rigorously test the relationship between traits proposed to be related to introduced/invasive species and introduced/invasive species’ high performance, specifically in the form of higher survival, higher fecundity and higher competitive ability.

Our findings also carry an important message for managers – as previously suggested we must not assume that all introduced species are super-plants [70-72]. Although some introduced species are clearly extremely successful in their new ranges (e.g. Lythrum salicaria has high recruitment success in North America [73], as does darwinii in

New Zealand [74]), both introduced and native species have a wide range of recruitment success, and on average introduced species’ survival is no better than that of native species.

If introduced species have no advantage during recruitment, then it is worth asking where in the life cycle they do out-perform native species. Theory suggests that introduced species, particularly invasive ones, will have advantages over native species in the form of

85 Recruitment of introduced and native species

abundant seed output [3], superior seed dispersal [72,75], faster and higher proportions of germination under a wider range of environmental conditions [5] and higher seedling survival and establishment [3]. Many empirical studies show that introduced species germinate faster [20-23,33,34] and under a wider range of conditions [24,28] than their native counterparts. However, the evidence does not always support this view, [29,30], and even if introduced species do have faster germination, under a wider range of conditions, these advantages may not necessarily translate to higher survival through the early stages of recruitment (Fig. 1). Introduced species produce as much as 6.7 times more seeds per plant per year than do native species [76], but a compilation of published data found no clear difference in seed dispersal distance between native and introduced species [69]. That is, higher seed production might be the only life history stage where introduced species have a general advantage over native species. However not all of the seeds plants produce are viable [77-79]. Another important topic for future research is to determine whether introduced species produce a higher proportion of viable seeds than do native species.

Introduced species’ higher seed production could give them a numerical advantage over native species during the early stages of recruitment. For instance, a study by Ramula et al.

[80] showed that introduced species’ population growth rate is on average 9.4 times higher compared to that of native species. This is roughly consistent with the 6.7 times higher seed production [76] being the only detectable advantage of introduced species. However,

Ramula et al.[80] also highlighted the tendency for ecologists to measure population growth rates of introduced species during the exponential growth phase, but to measure native species’ population growth rates when they are stable or declining. Higher seed

86 Recruitment of introduced and native species

production could also positively affect invasive species’ capacity to form a persistent seed bank in their introduced ranges [81]. It is worth asking: when during the introduction process does the effect of higher seed production start, how long does it last and does seed production differ between introduced and native populations of the same species?

Although we have by far the largest dataset ever compiled on recruitment success of plant species, there were far fewer data available for survival from germination to first reproduction than for survival through germination or early seedling survival. This is most likely because following the fate of seeds beyond the early stages of survival particularly in long lived species is extremely difficult. In our study we found no significant difference between introduced (n = 18) and native species’ (n = 24) survival from germination to first reproduction. Given our modest sample size, a lack of statistical power might have contributed to the non-significant difference between introduced and native species’ survival from germination to first reproduction (Fig. 1C). However, only 28 replicates would be needed to accurately (90% accuracy) detect a 5% difference between introduced and native species’ survival from germination to first reproduction (using a power test;

[82]). That is, any difference between introduced and native species’ survival from germination to first reproduction would have to be fairly small in order to not have been detected in our study.

In summary, we have shown that on average, introduced species do not have higher recruitment success than do native species. These results challenge our traditional

87 Recruitment of introduced and native species

understanding of recruitment as a key driver of introduced species success. To understand the driving factors behind introduced species’ success, we need to quantify the relevance of traits and mechanisms proposed to be of importance to introduced species under natural conditions. It is time for invasion ecology to go through a careful synthesis and integration of the theories and empirical information and to determine whether introduced species actually do have, or should even be expected to have, lasting advantages over native species.

Acknowledgements

H.F.M. wishes to thank Rhiannon Dalrymple and Andrew Letten for their constructive comments on previous versions of the manuscript. We are grateful to Timothy Hitchcock for continuous longevity data. H.F.M. was supported by a scholarship from the Evolution &

Ecology Research Centre at UNSW. A.T.M. was supported by funding from the Australian

Research Council (DP 0984222). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1. Moles AT, Westoby M (2006) Seed size and plant strategy across the whole life cycle.

Oikos 113: 91-105.

88 Recruitment of introduced and native species

2. Kitajima K, Fenner M (2000) Ecology of seedling regeneration. In: Fenner M, editor.

Seeds: The ecology of regeneration in plant communities. New York: CABI

Publishing. pp. 331-359.

3. Pyšek P, Richardson DM (2007) Traits associated with invasiveness in alien plants:

where do we stand? In: Nentwig W, editor. Biological Invasions, Ecological Studies

Vol 193. Berlin: Springer-Verlag. pp. 97-125.

4. Rejmanek M, Richardson DM (1996) What attributes make some plant species more

invasive? Ecology 78: 171-181.

5. Colautti RI, Grigorovich IA, MacIsaac HJ (2006) Propagule pressure: a null model for

biological invasions. Biol Invasions 8: 1023-1037.

6. Gurevitch J, Fox GA, Wardle GM, Inderjit, Taub D (2011) Emergent insights from the

synthesis of conceptual frameworks for biological invasions. Ecol Lett 14: 407-418.

7. Keane RM, Crawley MJ (2002) Exotic plant invasions and the enemy release hypothesis.

Trends Ecol Evol 17: 164-170.

8. Chun YJ, van Kleunen M, Dawson W (2010) The role of enemy release, tolerance and

resistance in plant invasions: linking damage to performance. Ecol Lett 13: 937-

946.

9. Colautti RI, Grigorovich IA, MacIsaac HJ (2004) Is invasion success explained by the

enemy release hypothesis? Ecol Lett 7: 721-733.

10. Liu H, Stiling P (2006) Testing the enemy release hypothesis: a review and meta-

analysis. Biol Invasions 8: 1535-1545.

11. Daehler CC (2003) Performance comparisons of co-occurring native and alien invasive

plants: implications for conservation and restoration. Annu Rev Ecol Evol Syst 34:

183-211.

89 Recruitment of introduced and native species

12. Davidson AM, Jennions M, Nicotra AB (2011) Do invasive species show higher

phenotypic plasticity than native species and, if so, is it adaptive? A meta-analysis.

Ecol Lett 14: 419-431.

13. Palacio-López K, Gianoli E (2011) Invasive plants do not display greater phenotypic

plasticity than their native or non-invasive counterparts: a meta-analysis. Oikos 120:

1393-1401.

14. Dawson W, Rohr RP, van Kleunen M, Fischer M (2012) Alien plant species with a

wider global distribution are better able to capitalize on increased resource

availability. New Phytol 194: 859-867.

15. Sakai AK, Allendorf FW, Holt JS, Lodge DM, Molofsky J, et al. (2001) The population

biology of invasive species. Annu Rev Ecol Evol Syst 32: 305-332.

16. van Kleunen M, Weber E, Fischer M (2010) A meta-analysis of trait differences

between invasive and non-invasive plant species. Ecol Lett 13: 235-245.

17. Bellingham PJ, Duncan RP, Lee WG, Buxton RP (2004) Seedling growth rate and

survival do not predict invasiveness in naturalized woody plants in New Zealand.

Oikos 106: 308-316.

18. Grotkopp E, Rejmánek M (2007) High seedling relative growth rate and specific leaf

area are traits of invasive species: Philogenetically independent contrasts of woody

angiosperms. Am J Bot 94: 526-532.

19. Flory SL, Clay K (2009) Effects of roads and forest successional age on experimental

plant invasions. Biol Conserv 142: 2531-2537.

20. van Kleunen M, Johnson SD (2007) South African Iridaceae with rapid and profuse

seedling emergence are more likely to become naturalized in other regions. J Ecol

95: 674-681.

90 Recruitment of introduced and native species

21. Chrobock T, Kempel A, Fischer M, van Kleunen M (2011) Introduction bias: cultivated

alien plant species germinate faster and more abundantly than native species in

Switzerland. Basic Appl Ecol 12: 244-250.

22. Honig MA, Cowling RM, Richardson DM (1992) The invasive potential of Australian

banksias in South African fynbos: A comparison of the reproductive potential of

Banksia ericifolia and Leucadendron laureolum. Austral Ecol 17: 305-314.

23. Klink CA (1996) Germination and seedling establishment of two native and one

invading African grass species in the Brazilian cerrado. J Trop Ecol 12: 139-147.

24. Smith MA, Bell DT, Loneragan WA (1999) Comparative seed germination ecology of

Austrostipa compressa and Ehrharta calycina (Poaceae) in a Western Australian

Banksia woodland. Austral Ecol 24: 35-42.

25. Siemann E, Rogers WE (2003) Herbivory, disease, recruitment limitation, and success

of alien and native tree species. Ecology 84: 1489-1505.

26. Davis NE, Forsyth DM, Coulson G (2009) Facilitative interactions between an exotic

mammal and native and exotic plants: hog deer (Axis porcinus) as seed dispersers in

south-eastern Australia. Biol Invasions 12: 1079-1092.

27. Vilà M, D'Antonio C (1998) Fruit choice and seed dispersal of invasive vs. non-

invasive Carpobrotus (Aizoaceae) in Coastal . Ecology 79: 1053-1060.

28. Cervera JC, Parra-Tabla V (2009) Seed germination and seedling survival traits of

invasive and non-invasive congeneric Ruellia species (Acanthaceae) in Yucatan,

Mexico. Plant Ecol 205: 285-293.

29. Mandák B (2003) Germination requirements of invasive and non-invasive Atriplex

species: a comparative study. Flora 198: 45-54.

91 Recruitment of introduced and native species

30. Frenot Y, Gloaguen J-C (1994) Reproductive performance of native and alien

colonizing phanerogams on a glacier foreland, Iles Kerguelen. Polar Biol 14: 473-

481.

31. Van Clef M, Stiles EW (2001) Seed longevity in three pairs of native and non-native

congeners: Assessing invasive potential. Northeast Nat 8: 301-310.

32. Sher AA, Marshall DL (2003) Seedling competition between native Populus deltoides

(Salicaceae) and exotic Tamarix ramosissima (Tamaricaceae) across water regimes

and substrate types. Am J Bot 90: 413-422.

33. Gerlach JD, Rice KJ (2003) Testing life history correlates of invasiveness using

congeneric plant species. Ecology 13: 167-179.

34. Schlaepfer DR, Glättli M, Fischer M, van Kleunen M (2010) A multi-species

experiment in their native range indicates pre-adaptation of invasive alien plant

species. New Phytol 185: 1087-1099.

35. Yamashita N, Tanaka N, Hoshi Y, Kushima H, Kamo K (2003) Seed and seedling

demography of invasive and native trees of subtropical Pacific islands. J Veg Sci

14: 15-24.

36. Horton JL, Clark JL (2001) Water table decline alters growth and survival of Salix

gooddingii and Tamarix chinensis seedlings. For Ecol Manage 140: 239-247.

37. Corbin JD, D’Antonio CM (2010) Not novel, just better: competition between native

and non-native plants in California grasslands that share species traits. Plant Ecol

209: 71-81.

38. Garcia-Serrano H, Escarré J, Sans FX (2004) Factors that limit the emergence and

establishment of the related aliens Senenecio inaequidens and Senecio pterophorus

92 Recruitment of introduced and native species

and the native Senecio malacitanus in Mediterranean climate. Can J Bot 82: 1346-

1355.

39. Morrison JA, Mauck K (2007) Experimental field comparison of native and non-native

maple seedlings: natural enemies, ecophysiology, growth and survival. J Ecol 95:

1036-1049.

40. Schumacher E, Kueffer C, Edwards PJ, Dietz H (2009) Influence of light and nutrient

conditions on seedling growth of native and invasive trees in the Seychelles. Biol

Invasions 11: 1941-1954.

41. Sher AA, Marshall DL, Taylor JP (2002) Establishment patterns of native Populus and

Salix in the presence of invasive non-native Tamarix. Ecol Appl 12: 760-772.

42. Sanford NL, Harrington RA, Fownes JH (2003) Survival and growth of native and alien

woody seedlings in open and understory environments. For Ecol Manage 183: 377-

385.

43. McAlpine KG, Jesson LK (2008) Linking seed dispersal, germination and seedling

recruitment in the invasive species Berberis darwinii (Darwin’s barberry). Plant

Ecol 197: 119-129.

44. Moles AT, Westoby M (2004) Seedling survival and seed size: a synthesis of the

literature. J Ecol 92: 372-383.

45. Westoby M, Falster DS, Moles AT, Vesk PA, Wright IJ (2002) Plant ecological

strategies: some leading dimensions of variation between species. Annu Rev Ecol

Evol Syst 33: 125-159.

46. Radford IJ, Cousens D (2000) Invasiveness and comparative life-history traits of exotic

and indigenous Senecio species in Australia. Oecologia 125: 531-542.

93 Recruitment of introduced and native species

47. Muth NZ, Pigliucci M (2007) Implementation of a novel framework for assessing

species plasticity in biological invasions: responses of Centaurea and Crepis to

phosphorus and water availability. J Ecol 95: 1001-1013.

48. Thébaud C, Finzi AC, Affre L, Debussche M, Escarre J (1996) Assessing why two

introduced Conyza differ in their ability to invade Mediterranean old fields. Ecology

77: 791-804.

49. Grigulis K, Sheppard AW, Ash JE, Groves RH (2001) The comparative demography of

the pasture weed Echium plantagineum between its native and invaded ranges. J

Appl Ecol 38: 281-290.

50. Rees M, Paynter Q (1997) Biological control of Scotch broom: modelling the

determinants of abundance and the potential of impact introduced insect herbivores.

J Appl Ecol 34: 1203-1221.

51. Cadotte MW, Murray BR, Lovett-Doust J (2006) Ecological patterns and biological

invasions: Using regional species inventories in macroecology. Biol Invasions 8:

809-821.

52. Tummers B (2006) DataThief III. Available: http://datathief.org/. Accessed: 2011

November.

53. Randall RP (2007) Global Compedium of Weeds. Melbourne, Australia: RG and FJ

Richardson. 906 p.

54. Moles AT, Ackerly DD, Webb CO, Tweddle JC, Dickie JB, et al. (2005) Factors that

shape seed mass evolution. Proc Natl Acad Sci U S A 102: 10540-10544.

55. Royal Botanical Gardens Kew (2008) Seed Information Database (SID). Version 7.1.

Available:http://data.kew.org/sid/ Accessed 2012 April.

94 Recruitment of introduced and native species

56. Warton DI, Hui FKC (2011) The arcsine is asine: the analysis of proportions in

ecology. Ecology 92: 3-10.

57. R Core Team (2013) R: a language and environment for statistical computing.

Available: http://www.r-project.org. Accessed: 2013 June.

58. Richardson DM, Pyšek P, Rejmánek M, Barbour MG, Dane F, et al. (2000)

Naturalization and invasion of alien plants: concepts and definitions. Divers Distrib

6: 93-107.

59. Garland T, Harvey PH, Ives AR (1992) Procedures for the analysis of comparative data

using phylogenetically independent contrasts. Syst Biol 41: 18-32.

60. Webb CO, Donoghue MJ (2005) Phylomatic: tree assembly for applied phylogenetics.

Mol Ecol Notes 5: 181-183.

61. Stevens PF (2001 onwards) Angiosperm Phylogeny Website.

Available:http://www.mobot.org/MOBOT/research/APweb/. Accessed 2012 April.

62. Webb CO, Ackerly DD, Kembel SW (2008) Phylocom: software for the analysis of

phylogenetic community structure and trait evolution. Bioinformatics 24: 2098-

2100.

63. Moles AT, Warton DI, Warman L, Swenson NG, Laffan SW, et al. (2009) Global

patterns in plant height. J Ecol 97: 923-932.

64. Wright IJ, Reich PB, Westoby M, Ackerly DD, Baruch Z, et al. (2004) The worldwide

leaf economics spectrum. Nature 12: 821-827.

65. Moles AT, Wallis IR, Foley WJ, Warton DI, Stegen JC, et al. (2011) Putting plant

resistance traits on the map: a test of the idea that plants are better defended at lower

latitudes. New Phytol 191: 777-788.

95 Recruitment of introduced and native species

66. Hedges LV, Gurevitch J, Curtis PS (1999) The meta-analysis of response ratios in

Ecology. Ecology 80: 1150-1156.

67. Vetter D, Rucker G, Storch I (2013) Meta-analysis: A need for well-defined usage in

ecology and conservation biology. Ecosphere 4: 1-24.

68. Davidson AM, Nicotra AB (2012) Beware: alien invasion. Where to next for an

understanding of weed ecology? New Phytol 194: 602-605.

69. Flores-Moreno H, Thomson FJ, Warton DI, Moles AT (2013) Are introduced species

better dispersers than native species? A broad comparative study of seed dispersal

distance. PLoS One 8: e68541.

70. Thompson K, Hodgson JG, Rich TCG (1995) Native and alien invasive plants: more of

the same? Ecography 18: 390-402.

71. Williamson MH (1999) Invasions. Ecography 5: 5-12.

72. Thompson K, Davis MA (2011) Why research on traits of invasive plants tells us very

little. Trends Ecol Evol 26: 155-156.

73. Blossey B, Notzold R (1995) Evolution of increased competitive ability in invasive

nonindigenous plants: A hypothesis. J Ecol 83: 887-887.

74. Allen RB (1991) A preliminary assessment of the establishment and persistence of

Berberis darwinii Hook., a naturalised shrub in secondary vegetation near Dunedin,

New Zealand. NZ J Bot 29: 353-360.

75. Jongejans E, Shea K, Skarpaas O, Kelly D, Sheppard AW, et al. (2008) Dispersal and

demography contributions to population spread of Carduus nutans in its native and

invaded ranges. J Ecol 96: 687-697.

76. Mason RAB, Cooke J, Moles AT, Leishman MR (2008) Reproductive output of

invasive versus native plants. Glob Ecol Biogeogr 17: 633-640.

96 Recruitment of introduced and native species

77. Bierzychudek P (1981) Pollinator limitation of plant reproductive effort. Am Nat 117:

838-840.

78. Cavers PB (1983) Seed demography. Can J Bot 61: 3578-3590.

79. Stephenson AG (1981) and fruit abortion: proximate causes and ultimate

functions. Annu Rev Ecol Evol Syst 12: 253-279.

80. Ramula S, Knight TM, Burns JH, Buckley YM (2008) General guidelines for invasive

plant management based on comparative demography of invasive and native plant

populations. J Appl Ecol 45: 1124-1133.

81. Gioria M, Pyšek P, Moracova L (2012) Soil seed banks in plant invasions : promoting

species invasiveness and long-term impact on plant community dynamics. Preslia

84: 327-350.

82. Sokal RR, Rohlf FJ (1995) Biometry: the principles and practices of stastistics in

biological research. New York: W.H. Freeman. 887-887 p.

Supporting information

Table S1. Comparison of introduced and native species’ early seedling survival and survival from germination to first reproduction, excluding biennial species.

97 Recruitment of introduced and native species

Table S2. Comparison of introduced and native species’ recruitment success once the effect of lifespan categories and seed mass have been accounted for.

Table S3. Comparison of introduced and native species recruitment success after accounting for seed mass, lifespan categories, and when including a random effect for site.

Table S4. Phylogenetic independent contrast of introduced and native species recruitment success.

Table S5. Comparison of introduced and native species’ recruitment success once the effect of life form (woody/non-woody) has been taken into account.

Table S6. Comparison of introduced and native species’ recruitment success once the effect of continuous longevity has been accounted for.

98 Recruitment of introduced and native species

Table S1. Comparison of introduced and native species’ early seedling survival and seedling survival to reproduction excluding data for biennial species.

1) EARLY SEEDLING SURVIVAL (ONE WEEK AFTER GERMINATION) AND

LONGEVITY

Term Sum of squares d.f. P

Intercept 20.85 1 0.02

Species' status 10.42 1 0.10

Lifespan categories 29.39 1 0.006

Species status × Lifespan categories 13.84 1 0.06

Residuals 535.51 142

2) SEEDLING SURVIVAL TO REPRODUCTION AND LONGEVITY

Term Sum of squares d.f. P

Intercept 33.17 1 0.03

Species' status 0.54 1 0.78

Lifespan categories 2.5 1 0.55

Species status × Lifespan categories 8.87 1 0.26

Residuals 232.31 34

3) EARLY SEEDLING SURVIVAL (ONE WEEK AFTER GERMINATION),

LONGEVITY AND SEED MASS

99 Recruitment of introduced and native species

Term Sum of squares d.f. P

Intercept 13.5 1 0.07

Species' status 4.88 1 0.28

Lifespan categories 30.96 1 0.007

Seed mass 0.74 1 0.67

Species status × Lifespan categories 11.25 1 0.10

Species status × Seed mass 0.07 1 0.90

Seed mass × Lifespan categories 2.56 1 0.43

Species status × Lifespan categories ×

Seed mass 0.07 1 0.90

Residuals 456.39 111

100 Recruitment of introduced and native species

Table S2. Comparison of introduced and native species’ recruitment success once the effect of lifespan categories and seed mass have been accounted for.

Longevity could affect the recruitment survival of introduced and native species. In order to control for this confounding factor we ran a linear model were our predictor variables where lifespan categories (annual, biennial and perennial), species’ status (introduced and native) and the interaction between lifespan categories and status, and our response variables were survival to germination, early seedling survival (one week survival after emergence) and survival from germination to first reproduction.

Next, to account for the effect of seed mass on early seedling survival we ran a linear model where the predictor variables were seed mass, species status and their interaction and the response variable was early seedling survival.

Finally we ran a linear model to account for the effect of seed mass, lifespan categories and their interaction on the early seedling survival of introduced and native species. In this case the predictor variables were seed mass, lifespan categories and their interactions and the response variable was early seedling survival.

All analyses were ran on logit-transform survival data and log-transform seed mass data.

101 Recruitment of introduced and native species

1) SURVIVAL THROUGH GERMINATION AND LIFESPAN CATEGORIES

Term Sum of squares d.f. P

Intercept 35.86 1 0.003

Species’ status 2.32 1 0.89

Lifespan categories 6.34 2 0.8

Species’ status × Lifespan categories 0.34 2 0.84

Residuals 1098.17 261

2) EARLY SEEDLING SURVIVAL (ONE WEEK AFTER GERMINATION) AND

LIFESPAN CATEGORIES

Term Sum of squares d.f. P

Intercept 20.1 1 0.03

Species’ status 8.74 1 0.14

Lifespan categories 26.73 2 0.04

Species’ status × Lifespan categories 13.64 2 0.19

Residuals 480.19 120

3) SURVIVAL FROM GERMINATION TO FIRST REPRODUCTION AND LIFESPAN

CATEGORIES

Term Sum of squares d.f. P

Intercept 33.17 1 0.03

102 Recruitment of introduced and native species

Species’ status 0.54 1 0.77

Lifespan categories 4.9 2 0.69

Species’ status × Lifespan categories 8.93 2 0.51

Residuals 237.86 37

4) EARLY SEEDLING SURVIVAL (ONE WEEK AFTER GERMINATION) AND SEED

MASS

Term Sum of squares d.f. P

Intercept 284.11 1 <.0001

Species’ status 1.98 1 0.49

Seed mass 13.47 1 0.07

Species’ status × Seed mass 2.12 1 0.47

Residuals 494.68 122

5) EARLY SEEDLING SURVIVAL (ONE WEEK AFTER GERMINATION), SEED

MASS, AND LIFESPAN CATEGORIES

Term Sum of squares d.f. P

Intercept 13.5 1 0.07

Species’ status 4.88 1 0.27

Longevity 38.6 1 0.002

Seed mass 0.74 1 0.66

Species status × Lifespan categories 11.79 1 0.09

103 Recruitment of introduced and native species

Species status × Seed mass 0.07 1 0.9

Seed mass × Lifespan categories 2.57 1 0.42

Species’ status × Seed mass × Lifespan 0.07 1 0.9 categories

Residuals 512.86 130

104 Recruitment of introduced and native species

Table S3. Comparison of introduced and native species recruitment success after accounting for seed mass, lifespan categories, and when including a random effect for site.

To test the influence of site to site variation on the recruitment of introduced and native species we ran a linear mixed model in the same dataset used for our linear model were our predictor variables were species’ status, lifespan categories and a random effect for site and or dependent variables were survival through germination, early seedling survival and survival from germination to first reproduction. In the case of the early seedling survival we also included a predictor variable for seed mass because the significant relationship found between these two traits (see main text; [1]).

All analyses were ran on logit-transform survival data and log-transform seed mass data.

105 Recruitment of introduced and native species

1) SURVIVAL THROUGH GERMINATION, LIFESPAN CATEGORIES, AND

RANDOM EFFECT FOR SITE

Term Estimate P

Intercept -1.94 < 0.01

Species’ status 0.75 0.18

Lifespan categories (biennia)l 0.56 0.64

Lifespan categories (perennial) -0.08 0.89

Species status × Lifespan categories (biennial) -0.8 0.87

Species status × Lifespan categories (perennial) -0.79 0.4

2) EARLY SEEDLING SURVIVAL, SEEDMASS, LIFESPAN CATEGORIES, AND

RANDOM EFFECT FOR SITE

Term Estimate P

Intercept 3.35 < 0.01

Species’ status -0.26 0.47

Lifespan categories (biennial/perennial) 0.007 0.29

Seed mass -0.37 0.78

Species’ status × Lifespan categories (biennial/perennial) -1.01 0.13

Species’ status × Seed mass 0.54 0.87

Lifespan categories (biennial/perennial) × Seed mass 0.55 0.51

Species’ status × Lifespan categories (biennial/perennial) -0.28 0.96

× Seed mass

106 Recruitment of introduced and native species

3) SURVIVAL FROM GERMINATION TO FIRST REPRODUCTION, LIFESPAN

CATEGORIES, AND RANDOM EFFECT FOR SITE

Term Estimate P

Intercept -2.58 0.03

Species’ status 0.06 0.79

Lifespan categories (biennial/perennial) 0.69 0.49

Species’ status × Lifespan categories (biennial/perennial) -1.39 0.28

References

1. Moles AT, Westoby M (2004) Seedling survival and seed size: a synthesis of the literature. J

Ecol 92: 372-383.

107 Recruitment of introduced and native species

Table S4. Phylogenetic independent contrast of introduced and native species

recruitment success.

Trait number of contrasts Mean Standard P

deviation

Survival to germination (%) 265 -0.04 1.53 0.9

Early seedling survival (%) 150 0.51 2.56 0.31

Survival from germination to first 42 0.43 3.12 0.65 reproduction (%)

All analyses were done on logit-transform survival data.

108 Recruitment of introduced and native species

Table S5. Comparison of introduced and native species’ recruitment success once the effect of life form (woody/non-woody) has been taken into account.

1) PROPORTION OF NATIVE AND INTRODUCED SPECIES WITH

WOODY/NON-WOODY LIFE FORM

Species’ status

Life-form Introduced species Native species

Non-woody 42 (66.7 %) 118 (41.7 %)

Woody 21 (33.3 %) 165 (58.3 %)

2) SURVIVAL THROUGH GERMINATION AND LIFE FORM

Terms Estimate Standard error P

Intercept -2.48 0.39 < 0.0001

Species' status 0.32 0.45 0.47

Life form 0.45 0.62 0.47

Species' status × Life form -0.14 0.68 0.84

3) EARLY SEEDLING SURVIVAL (ONE WEEK AFTER GERMIANTION) AND

LIFE FORM

Terms Estimate Standard error P

Intercept 2.29 0.35 < 0.0001

Species' status 0.17 0.46 0.72

109 Recruitment of introduced and native species

Life form 1.57 0.68 0.02

Species' status × Life form -0.97 0.77 0.21

4) SURVIVAL FROM GERMINATION TO FIRST REPRODUCTION AND LIFE

FORM

Terms Estimate Standard error P

Intercept -1.59 0.71 0.03

Species' status -0.66 0.94 0.48

Life form -0.74 1.17 0.53

Species' status × Life form -0.94 1.58 0.56

110 Recruitment of introduced and native species

Table S6. Comparison of introduced and native species’ recruitment success once the effect of continuous longevity has been accounted for.

We collected maximum recorded longevity (continuous longevity) data from the global literature. Data were collected from a search performed in ISI web of knowledge using the search string: (“longevity” or “lifespan” or “maximum age”) and (“tree” or “shrub” or “herb” or “plant” or “grass”) not (“leaf lifespan” or “floral longevity” or “seed longevity”), only in fields (“plant sciences”, “environmental sciences”, “ecology”, “forestry”, “evolutionary biology”, “developmental biology” and “demography”). Only articles or reviews in English were considered. The search was performed the 5th of December of 2012 and contained information for papers published after 2001. Information on maximum longevity data from papers published before 2001 were sourced from Moles et al. [1]. For 17 species the data on maximum longevity was extracted from Ramula et al. [2]. Finally, the maximum longevity of three species was estimated using the species’ population matrices (following [2-4]). First, for each species, the fecundity values from the population matrices were set to zero. Then a population vector with one seedling and zero adults was multiplied by the population matrices until the summed probability of survival for all size classes summed up to 0.01. In total we were able to collect longevity data for 128 species, 23 introduced and 105 native. All longevity data were log10-transformed before analysis.

1) SURVIVAL THROUGH GERMINATION AND CONTINUOUS LONGEVITY

Terms Sum of squares d.f. P

Intercept 65.28 1 0.0004

111 Recruitment of introduced and native species

Species' status 16.32 1 0.07

Continuous longevity 15.77 1 0.07

Species' status × Continuous longevity 19.42 1 0.05

Residuals 471.33 97

2) EARLY SEEDLING SURVIVAL (ONE WEEK AFTER GERMINATION) AND

CONTINUOUS LONGEVITY

Terms Sum of squares d.f. P

Intercept 72.99 1 < 0.0001

Species' status 3.82 1 0.25

Longevity 5.35 1 0.18

Species' status × Continuous longevity 0.414 1 0.71

Residuals 146.789 51

3) EARLY SEEDLING SURVIVAL (ONE WEEK AFTER GERMINATION), SEED

MASS, AND CONTINUOUS LONGEVITY

Terms Sum of squares d.f. P

Intecept 65.14 1 <0.0001

Species' status 6.95 1 0.16

Continuous longevity 2.22 1 0.42

Seed mass 2.82 1 0.36

Species' status × Continuous longevity 0.03 1 0.93

Species' status × Seed mass 1.18 1 0.56

112 Recruitment of introduced and native species

Continuous longevity × Seed mass 0.22 1 0.8

Species' status × Continuous longevity ×

Seed mass 0.69 1 0.65

Residuals 140.92 42

4) SURVIVAL FROM GERMINATION TO FIRST REPRODUCTION AND

CONTINUOUS LONGEVITY

Terms Sum of squares d.f. P

Intercept 25.99 1 0.04

Species' status 1.15 1 0.65

Continuous longevity 0.7 1 0.73

Species' status × Continuous longevity 9.27 1 0.21

Residuals 157.15 28

References

1. Moles AT, Falster DS, Leishman MR, Westoby M (2004) Small-seeded species produce

more seeds per square metre of canopy per year , but not per individual per lifetime. J

Ecol 92: 384-396.

2. Ramula S, Knight TM, Burns JH, Buckley YM (2008) General guidelines for invasive

plant management based on comparative demography of invasive and native plant

populations. J Appl Ecol 45: 1124-1133.

3. Caswell H (2001) Matrix population models: construction, analysis, and interpretation.

Sunderland, MA, USA: Sinauer. 722-722 p.

113 Recruitment of introduced and native species

4. Forbis TA, Doak DF (2004) Seedling establishment and life history trade-offs in alpine

plants. Am J Bot 91: 1147-1153.

114

Chapter three

Nature vs. Nurture: Are introduced species successful because

of what they are, or where they are?

Habacuc Flores-Moreno, Fiona J. Thomson, Rhiannon L. Dalrymple & Angela T.

Moles

The study was conceived and designed by HFM. RLD and ATM. HFM performed the

experiments. HFM and FJT collated the data. HFM, RLD and ATM analysed the data.

The manuscript writing was led by HFM, but all co-authors contributed to it.

Abstract

A species that is native in one place can be a serious invader elsewhere in the world.

Studies seeking to understand the causes of invasiveness have either focused on describing traits related to invasive species (nature) or on the effect that range change

(nurture) exerts on species’ performance. However, little is known about the relative importance of nature and nurture on species’ level of invasiveness. We asked whether species with different levels of invasiveness differ in key life history characteristics. To

115 do this we compiled a dataset including two functional traits (plant height and seed mass) and four life history processes (survival through germination, early seedling survival, reproductive output and seed dispersal distance) for 1222 species from the global literature. First we investigated the effect of species’ characteristics (nature).

Species that have never established outside their native range (untransported species) are taller plants with heavier seeds than are invasive and naturalized species. Invasive species had higher seed production than did naturalized and untransported species, but there were no significant differences in seed dispersal, or survival. Next, we investigated the effect of range (nurture) by comparing the life history characteristics of invasive species in their introduced vs. native range. Invasive species had bigger seeds and higher seed production in their introduced range than in their home range, but showed no significant difference in survival or dispersal. Level of invasiveness and invasive species’ range explained up to 8 % and 12 % of the variation in the life history characteristics that we compared in our study, respectively. This shows that differences between invasive and non-invasive species are in part linked to both invasive species’ intrinsic characteristics (nature) and to the introduced range (nurture). Our results further the understanding of the basic ecological mechanisms behind invasive species’ success by pinpointing higher seed production as a key life history characteristic across invasive species around the world.

Introduction

There has been enormous interest in quantifying differences between invasive and non- invasive species (Daehler 2003, van Kleunen et al. 2010, Richardson and Pyšek 2012).

However, most studies trying to understand invasive species’ success focus either on

116 characterizing how species-specific traits enhance invasive species’ performance compared to non-invasive species (nature; e.g. Rejmanek and Richardson 1996,

Leishman et al. 2007, Burns et al. 2013) or on how trait variation between introduced compared to native ranges affects invasive species’ performance (nurture; e.g. Blossey and Notzold 1995, Buckley et al. 2003, Jongejans et al. 2008). This situation limits our knowledge about the relative importance that nature and nurture have on invasive species’ success. In this paper, our overall aim is to further our understanding of plant invasions by disentangling the effects of species’ ranges, and species’ levels of invasiveness at a broad scale under natural conditions.

Nature: effect of species’ traits and life history characteristics on invasiveness

Our first aim was to make a cross species comparison of seed mass, plant height, seed production, seed dispersal distance, germination success and early seedling survival of species with different level of invasiveness (A vs. (B and C) – Grey dashed box in

Figure 1). We chose these traits and life history characteristics because all of them are thought to have a substantial effect on species’ ecological strategy and/or fitness

(Westoby et al. 2002, Moles et al. 2004, Moles and Westoby 2004, Moles et al. 2005,

Moles et al. 2009, Thomson et al. 2011). Furthermore, many studies comparing morphological and functional traits of invasive species with naturalized or native species at local, regional and global scales have suggested that invasive species may disperse seeds further, have higher recruitment success and have higher reproductive output (Rejmanek and Richardson 1996, Grotkopp and Rejmánek 2007, Pyšek and

Richardson 2007, van Kleunen and Johnson 2007b, Richardson and Pyšek 2012).

Comparing the differences in fitness-related life history characteristics and traits

117 between species with different levels of invasiveness provides us with a baseline. This is our first aim.

Previous studies have described differences in plant traits between invasive and non- invasive species (summarized in: Daehler 2003, Pyšek and Richardson 2007, van

Kleunen et al. 2010, Richardson and Pyšek 2012). Plant height and seed mass are two ecologically important traits (Westoby et al. 2002, Moles et al. 2009) that impact seed production (Moles et al. 2004), dispersal (Thomson et al. 2011) and early seedling survival (Moles and Westoby 2004) of plant species. Differences in plant height and/or seed mass between species with different levels of invasiveness could lead to spurious differences in these life history characteristics. Therefore, our second aim was to ask whether species with different levels of invasiveness showed significant differences in seed production, seed dispersal distance, or early seedling survival once the effects of traits such as plant height and seed mass had been accounted for (A vs. (B and C) –

Grey dashed box in Figure 1).

Nurture: effect of the native range

The two comparisons (Aims 1 and 2) provide useful baseline information, but they include values for introduced species growing in both their native and introduced ranges

(A with (B and C) – Grey dashed box in Figure 1). That is, any differences could result from differences in the species’ levels of invasiveness, or from differences between the native and introduced range.

118 Theory suggests that higher performance of invasive species could result from intrinsic differences between invasive and non-invasive species already present before introduction (van Kleunen et al. 2007, Schlaepfer et al. 2010). Previous studies have compared characteristics of invasive and non-invasive species in their native range, generally focusing on certain taxa and/or region (Rejmanek and Richardson 1996, van

Kleunen and Johnson 2007b, van Kleunen et al. 2007, Schlaepfer et al. 2010, Castro-

Díez et al. 2011, Gallagher et al. 2011). This approach accurately distinguishes the characteristics of invasive species as it reduces the possible variation due to species having different regions of origin and/or across taxa variation (e.g. Rejmanek and

Richardson 1996, van Kleunen and Johnson 2007b, Castro-Díez et al. 2011). For instance, plant height is positively correlated with invasiveness in Australian Acacia species (Castro-Díez et al. 2011, Gallagher et al. 2011) and South African Iridaceae species (van Kleunen et al. 2007), but it is not correlated with invasiveness in Pinus species (Rejmanek and Richardson 1996). Similarly, seed mass is negatively correlated with invasiveness in Pinus species (Rejmanek and Richardson 1996) and European species introduced to North America (Schlaepfer et al. 2010), but not with South

African Iridaceae species (van Kleunen et al. 2007). Many of the characteristics compared in these studies are only examined in one study, some examples are: genome size (Gallagher et al. 2011), serotony (Rejmanek and Richardson 1996), and resprouting ability (Castro-Díez et al. 2011), and plasticity. This approach may prove challenging when looking for broad patterns of invasiveness; characteristics that are identified as important correlates of invasiveness for some taxa or regions, are not important in others. Furthermore, the fact that many characteristics compared are only examined in one study reduces our power to generalize across studies. Nobody has done a broad scale cross-species comparison of species with different levels of invasiveness only

119 including measurements taken on species growing in their native range (A vs. B Black box in Figure 1). This was our third aim. Determining whether differences between species with different levels of invasiveness remain significant, after accounting for the effect of range, will help us understand the relative importance of species’ intrinsic characteristic for their level of invasiveness.

Figure 1. Diagram of our comparisons between levels of invasiveness, when the species’ range (native and introduced) is considered. (A) represents untransported species growing in their native ranges. (B) represent species that have been recorded as introduced somewhere in the world (naturalized or invasive), but growing in their native ranges. (C) represents introduced species (invasive and naturalized) growing in their introduced ranges. Our first comparison was of species with different levels of invasiveness, regardless of range (gray box; A vs. B and C). Our second comparison was of species with different levels of invasiveness, but growing within their native ranges (solid black box; A vs. B). Our third comparison was of invasive species in their native vs. introduced ranges (black dashed box; B vs. C).

Nurture: effect of the introduced range

Evolutionary and ecological theory leads us to expect that invasive species will display higher fitness as a benefit of being introduced into a new range. Higher performance of

120 invasive species or differences in traits could arise as a result of the new selective forces and abiotic/biotic conditions that the species experience in their new range (increased competitive ability, less herbivory, more nutrients; Blossey and Notzold 1995). For instance, invasive species in their introduced range might undergo enemy release (Liu and Stiling 2006, but see Chun et al. 2010). This could lead to evolution of increased competitive ability (Blossey and Notzold 1995). Alternatively, invasive species could benefit from having higher plasticity levels (Davidson et al. 2011), or simply they could be better at capitalizing resources (Dawson et al. 2012).

Several studies have compared the traits of invasive species at the within species level

(e.g. Blossey and Notzold 1995, Buckley et al. 2003, Jongejans et al. 2008). However, it is unknown whether there is a general trend in the effect of range (native vs. introduced range) on the characteristics of invasive species at the cross-species level at a broad scale under natural conditions. Recently, a meta-analysis by Parker et al. (2013) showed that, on average, invasive species in their introduced range are larger, more fecund and more abundant than in their home range. However, Parker et al. (2013) grouped a wide variety of traits (37 different traits related to reproduction and 65 traits reflecting plant size) and sometimes even grouped together life history characteristics that are negatively correlated (e.g. seed mass and number of seeds per individual). Therefore, although Parker et al.’s (2013) analysis is broad, from its results we are unable to disentangle what reproductive and size related traits differ between ranges of invasive species. Our fourth aim was to do a broad scale cross species comparison of characteristics of invasive species in their home range with their characteristics in their introduced range (B vs. C – Black dashed box in Figure 1). Determining whether or not there are meaningful differences in key traits (seed mass and plant height) and vital life

121 histories characteristics (dispersal, survival and seed production) between invasive species in their native and introduced ranges will help us identify how much invasive species fitness/performance is modulated by the environment. This comparison will help us to uncover whether or not invasive species in their introduced range tend to allocate resources to specific life history characteristics or if the shift in resource allocation in the introduced range is idiosyncratic to the species and/or environmental conditions that they experience.

In summary in this paper we are going to address four aims:

1) To determine whether species with different levels of invasiveness differ in their

traits and life history characteristics.

2) To determine whether species with different levels of invasiveness differ in

survival through germination, early seedling survival, reproductive output and

seed dispersal distance once the effects of ecological important traits have been

accounted for.

3) To determine whether native species with different level of invasiveness differ

in their traits and life history characteristics.

4) To determine whether invasive species in the native range differ in traits and life

history characteristics to invasive species in the introduced range.

Methods

122 We started with the databases on seed production, seed dispersal, and recruitment of introduced and native species compiled by Mason et al. (2008), Flores-Moreno et al.

(2013) and Flores-Moreno and Moles respectively (unpublished). The backbone for these data sets comes from data collected mainly for native species by Moles and

Westoby (2004) for recruitment data, Moles et al. (2004) for seed production data and

Thomson et al. (2011) for seed dispersal data. Data on plant height and seed mass were extracted from the same paper as the seed production, seed dispersal and recruitment data; alternatively, plant height was sourced from Moles et al. (2009) and seed mass from Moles et al. (2005).

Where possible we extracted information on level of invasiveness from the same papers as seed production, dispersal distance and recruitment data. However, to have a more accurate overall measurement of a species’ invasiveness we cross-referenced this list with four other sources: 1) The Global Compendium of Weeds (Randall 2007a following , van Kleunen et al. 2007, Pyšek et al. 2009, Dawson et al. 2011, Parker et al.

2013), 2) the Germoplasm Resources Information Network (USDA ARS National

Genetic Resources Program 2013), 3) floras from environmental agencies (e.g. USDA

Plant Database, Flora of New Zealand, Australian Virtual Herbarium, Online Atlas of the British and Irish Flora), and 4) regional invasive species databases or studies (e.g.

Randall 2007b, CONABIO 2012, Pyšek et al. 2012, DAISIE 2013). Species that were not listed as introduced in any of the resources above were considered to be uniquely distributed in their region of origin, and are not known to be introduced or invasive anywhere in the world.

123 We use three categories of invasiveness: Invasive species have rapid population increase and rapid range expansion (sensu Richardson et al. 2000). Naturalized species establish self-replacing populations without range expansion (sensu Richardson et al. 2000). In this context, untransported species are native species uniquely distributed in their region of origin, and are not known to be introduced or invasive anywhere in the world.

From the 1222 species included in our database, 663 were classified as untransported,

125 as naturalized and 440 as invasive. The sample size for the life history characteristics included in each of our tests is reported under each subgraph inFigure 2.

Prior to analysis, we log10-transformed seed mass, plant height, number of seeds produced per individual per year, and mean and maximum dispersal distance. We logit- transformed survival through germination, and early seedling survival data. For the survival data, we added or subtracted the smallest non-zero value for species with survival values equal to one or zero (Warton and Hui 2011). All averages presented throughout our paper are geometric means.

First we tested whether species with different invasiveness level differ in plant height, seed mass, seed production, seed dispersal distance (mean and maximum), survival through germination, and early seedling survival. To do this we ran ANOVA tests in which species’ invasiveness level (i.e. invasive, naturalized and untransported) were the predictor variable and either the traits or life history characteristics of interest (e.g. seed mass, dispersal distance or seed production) were the response variable. Then, whenever appropriate, we ran posthoc Tukey-Kramer HSD tests to disentangle significant differences between invasiveness levels.

124 Next, we asked whether species with different levels of invasiveness differed in seed production, dispersal distance, and early seedling survival after accounting for the effect of ecologically relevant traits (i.e. seed mass and plant height). We ran these tests as linear models where the life history characteristic (e.g. seed production or dispersal distance) under comparison was our dependent variable and our predictor variables were the species’ level of invasiveness and the ecologically relevant traits (i.e. seed mass and/or plant height). For seed production and dispersal distance we included terms for seed mass and plant height, while for seedling survival we only included seed mass.

In the case of seed production we used plant height as a proxy of plant size because seed production is related to canopy area (Moles et al. 2004), and information for canopy area is very rare in the literature. Using a linear model and data for 99 species from

Moles et al. (2004) we asked whether plant height could be used as a proxy for canopy area. Our predictor variable was plant height and our response variable was canopy area. We found that plant height explains 74% of the variation on canopy area (P <

0.00001) across species. On average a plant 10 cm tall would have a canopy of 0.03 cm2, while a plant 85.4 m tall will have a canopy area around 7310 m2. Plant height and canopy area were log10-transform previous to any analysis.

We then asked whether differences between species with different levels of invasiveness could be due to a confounding effect from comparing species with different ranges. We re-ran the ANOVA tests on a subset of our database only containing species for which we have trait and life history characteristic data from their

125 native ranges. This subset of the data included 333 invasive, 102 naturalized and 653 untransported species.

Finally, to test how much of the variation in invasiveness is due to range we compared the characteristics of invasive species in their native range and invasive species in their introduced range. We re-ran the ANOVA tests on a subset of our dataset that contained data for 333 invasive species on their home range vs. 126 invasive species in their introduced range. All analyses were performed in R 2.15.3 (R Core Team 2013).

Results

Nature: effect of species’ traits and life history characteristics on invasiveness

There were significant differences in plant height, seed mass and seed production between species with different levels of invasiveness (P < 0.01 in all analyses; Fig. 2 A-

C). On average, untransported species were significantly taller plants (geometric mean =

1.9 m) with heavier seeds (8 mg) compared to invasive (1.5 m; 2.1 mg) and naturalized species (1.2 m; 1.9 mg; Fig. 2 A and B). Invasive species had significantly higher seed production (940 seeds.individual-1.yr-1) than untransported species (228 seeds.individual-1.yr-1; P < 0.0001), but not compared to naturalized species (420 seeds.individual-1.yr-1; P = 0.24; Fig. 2C; Supporting information 2).

126 Species with different levels of invasiveness did not significantly differ in mean or maximum dispersal distance (P = 0.20 and 0.62 respectively; Fig. 2 D and E), survival through germination (P = 0.32; Fig. 2 F) or early seedling survival (P = 0.40; Fig. 2 G;

Supporting information 2).

Incorporating terms for plant height, seed mass and/or longevity did not qualitatively affect any of our results. Differences in seed production per individual per year between species with different invasiveness levels remained significant (P = 0.0001; Supporting information 3), while differences in mean and maximum dispersal distance and early seedling survival remained non-significant (all P > 0.2; Supporting information 3).

127 -1 -1 Seeds.individual .year Seed mass [log scale] (mg) [log scale] Maximum dispersal distance Mean dispersal distance 10000000 Early seedling survival Survival trough germination 10000000 1000000 1000000

100000 Plant height (m) [log scale] (m) [log scale] 100000

(%) [logit scale] (%) [logit-scale] 0.0001 10000 10000 10000 10000 99.99 99.99 0.001 0.001 1000 1000 1000 (m) [log scale] 1000 99.9 99.9 0.01 0.01 0.01 0.01 100 100 100 100 100 0.1 0.1 0.1 0.1 0.1 10 30 50 70 90 99 10 30 50 70 90 99 10 10 10 10 10 1 1 1 1 1 1 1 D) G) F) B) E) C) A) naieNtrlzdUntransported Naturalized Invasive Species by level ofinvasiveness Species N = 114 N = 47 N = 244 N= 47 N= 114 N = N = 60 N = 14 N = 48 N= 14 N= 60 N = 138 N= 24 N= 94 N = 16 N= 72 N = N = 459 N = 116 N = 653 N= 116 N= 459 N = N = 442 N = 106 N = 523 N= 106 N= 442 N = N = 87 N = 29 N = 127 N= 29 N= 87 N = A A A AB A A P P P = 0.006 = 150 < 0.0001 < < 0.0001 < P P P P B B B = 0.20 = = 0.40 = 0.32 = = 0.62 = 128 Native species by level of invasiveness species Native J) K) M) N) L) I) H) naieNtrlzdUntransported Naturalized Invasive N = 24 N = 10 N = 48 N= 10 N= 24 N = N = 68 N = 41 N = 244 N= 41 N= 68 N = N = 44 N = 18 N = 108 N= 18 N= 44 N = 150 N= 12 N= 36 N = N = 333 N = 102 N = 653 N= 102 N= 333 N = N = 43 N = 24 N = 127 N= 24 N= 43 N = N = 329 N = 95 N = 523 N= 95 N= 329 N = A A A AB A A P P < 0.0001 < P P B P = 0.006 = B B P P = 0.02 = = 0.24 = = 0.96 = = 0.5 = 0.1 = U) T) R) Q) S) O) P) Invasive species by range Invasive species Native N = 24 N = 36 N = N = 45 N = N = 68 N = N = 333N = N = 43 N = N = 329N = Introduced N = 36 N = N = 44 N = N = 36 N = 49 N = N = 46 N = N = 113N = N = 126N = P P < 0.0001 < P P P P = 0.004 = P = 0.44 = = 0.53 = 0.48 = 0.47 = = 0.53 =

Figure 2. Comparison of the traits and life history characteristics of species by their level of invasiveness, and of invasive species by range (native and introduced). A-F is the comparison of native and introduced species by their level of invasiveness (invasive, naturalized and untransported species). H-N is the comparison of only native species by level of invasiveness. O-U is the comparison of invasive species in their native range and invasive species in their introduced range. Black dashed lines represent mean values. The boxes represent the 25th, 50th and 75th percentiles. Whiskers represent the 10th and 90th percentiles, outliers are represented as points. Sample sizes are number of species. Letters above boxes represent levels of invasiveness with significantly (P < 0.05) different mean traits or life history characteristics values.

129

Nurture: effect of the native range

Our results were not qualitatively affected when we compared native species with different level of invasiveness. Seed mass, plant height and seed production remained significantly different between species with different levels of invasiveness (invasive, naturalized and untransported), growing in their native range (P < 0.0001, P = 0.006; P

= 0.02, respectively; Fig. 2 H-J). Individually, seed mass, plant height and seed production accounted for 8 %, 0.9 % and 2 % of the variation in native species’ level of invasiveness (Supporting information 3). As in previous tests, there were no significant differences in mean dispersal distance (P = 0.1; Fig. 2 K), maximum dispersal distance

(P = 0.96; Fig. 2 L), germination (P = 0.5; Fig. 2 M) or early seedling survival (P =

0.24; Fig. 2 N; Supporting information 2) between species with different levels of invasiveness, growing in their native ranges. Early seedling survival accounted for 4 % of the variation on level of invasiveness, while mean dispersal distance accounted for 2

% of the variation. Maximum dispersal distance and germination accounted for < 0.7 % of variation on native species’ level of invasiveness (Supporting information 2).

Nurture: effect of the introduced range

When we compared invasive species in their introduced range vs. invasive species in their native range the results were qualitatively similar to previous comparisons.

Invasive species in their native range and invasive species in their introduced range were significantly different in seed mass and seed production (P < 0.0001 and 0.004;

Fig. 2 O and Q), but not in plant height (P = 0.44; Fig. 2 P). On average, invasive species in their introduced range had bigger seeds and higher seed production (4.63 mg;

130 2194 seeds.individual-1.yr-1) than did invasive species in their native range (1.56 mg;

530 seeds.individual-1.yr-1; Fig. 2 O and Q). The average total mass of yearly seed production, that is the product of the seed mass by the seed production per individual per year (seed mass × seeds.individual-1.yr-1), of invasive species in their introduced range was almost 200 times higher than the total seed production of invasive species in their native range (P = 0.0005). Seed mass, seed production and yearly seed production accounted for 4 %, 7 % and 12 % of the variation on invasive species’ range, respectively (Supporting information 3). Invasive species in their native range and invasive species in their introduced range did not significantly differ in mean dispersal distance (P = 0.47; Fig. 2 R), maximum dispersal distance (P = 0.53; Fig. 2 S), germination (P = 0.48; Fig. 2 T) or early seedling survival (P = 0.53; Fig. 2 U;

Supporting information 3). These characteristics accounted for < 0.73 % of the variation on invasive species’ range (Supporting information 3).

Discussion

Our most important finding is that differences between species with different levels of invasiveness are due to both differences between introduced and native ranges (nurture) and differences in the life history characteristics (nature) of the species. Overall, level of invasiveness explained between 0.04 to 8 % of variation on the traits and life history characteristics that we compared, while invasive species’ range (introduced vs. native) explained between 0.1 and 12 % of variation. This finding supports the idea that across invasive species there are some common life history characteristics (Kolar and Lodge

2001, Pyšek and Richardson 2007, Richardson and Pyšek 2012). However, for the majority of the life history characteristics that we compared (seed dispersal, survival through germination and early seedling survival) we found no significant differences between levels of invasiveness. Furthermore level of invasiveness explains less than 0.1

131 % of the variation in seed dispersal distance and recruitment success. These results suggest that the variation in certain life history characteristics might be more dependent on the environmental conditions that invasive species experience that on a species’ level of invasiveness. Taken together our results show that invasive species life histories are complex, and that invasive species do not out-perform untransported and naturalized species in all life history characteristics.

High seed production has been nominated as an important determinant of invasiveness

(Pyšek and Richardson 2007, Richardson and Pyšek 2012). Mason et al. (2008) showed that invasive species produce approximately seven times more seeds than native species

(invasive and non-invasive). However, until now it was unknown whether seed production differed between invasive, naturalized and untransported species. We found that invasive species produced two times more seeds than did naturalized species, and four times more seeds than did untransported species. An important future direction will be to determine how invasive species achieve higher seed production. Invasive species’ higher seed production could be explained by higher pollinator visitation or less accessory costs to seed production (e.g. less flower or fruit abortion, less costs of packing structures). For instance, a meta-analyses by Morales and Traveset (2009) showed a decline in pollinators visitation and native plants reproductive success in the presence of invasive plant species. Similarly, the invasive species Melaleuca quinquenervia has less aborted capsules in its’ introduced range (Rayamajhi et al.

2002). Invasive species’ capacity for self-compatibility or inbreeding (van Kleunen and

Johnson 2007a, Küster et al. 2009, but see Knapp and Kühn 2012) could also positively impact their seed production.

132 It has been suggested that higher seed production could positively affect the dispersal and/or survival of invasive species (Rejmánek 1996, Rejmanek and Richardson 1996,

Mason et al. 2008). Nevertheless, our results show that this is not the case; higher seed production of invasive species does not translate to higher dispersal or survival (survival through germination or early seedling survival). Studies by Ramula et al. (2008) and

Burns et al. (2013) showed that invasive species have a strategy more reliant on fecundity and growth than on survival than do native or non-invasive introduced species. However, Flores-Moreno et al. (2013; chapter one of this thesis) showed that introduced and native species do not differ in dispersal distance. One possible explanation of this result is that invasive species are less well-adapted to the local conditions in their new environment. For instance, a study by a Buswell et al. (2011) shows that the direction of the change in invasive species’ traits through time (leaf area, plant height and leaf shape) varies depending on the species’ identity. Introduced species often occupy different climatic conditions in their new ranges (Broennimann et al. 2007, Fitzpatrick et al. 2007, Gallagher et al. 2010, but see Petitpierre et al. 2012).

Alternatively, invasive species’ higher seed production might just come at the cost of reduced survival and/or dispersal. For instance, as a result of the trade off between seed production and seed mass. An important future direction for invasion biology is to determine under which circumstances higher seed production would have a positive or negative impact on different life stages.

We found that invasive species in their introduced range produce more and bigger seeds than do invasive species in their native range. This result suggests a positive effect of range on invasive species allocation of resources to reproduction, and is consistent with the findings of Parker et al. (2013), who showed that invasive species have higher

133 reproduction- and size-related traits in their introduced range than their native range. It is possible that invasive species might benefit from enemy release in their introduced range (Liu and Stiling 2006, Hawkes 2007 but see, Chun et al. 2010). Leishman et al.

(2007) showed that invasive species had higher foliar nitrogen, foliar phosphorus and light assimilation rate. That is, invasive species tend to have leaf traits related to high photosynthetic activity. This could enable invasive species to have rapid growth and/or more allocation of resources to reproduction. Leaf traits of invasive species could allow a “bang-bang” strategy (Falster and Westoby 2003) where invasive species would first maximize vegetative growth (Grotkopp et al. 2002, Grotkopp and Rejmánek 2007,

Dawson et al. 2011) and then switch resources to reproduction, possibly maximizing their reproductive output. Invasive species could also achieve higher seed production by higher increases in biomass in response to favourable conditions (Baker 1974, Davidson et al. 2011). For instance, Dawson et al. (2012) found that the most widespread introduced species tended to increase biomass with an increase in resource availability.

This could result in overall higher biomass and/or higher reproductive biomass.

However, it is not yet known how much each contributes to invasive species’ success.

This deserves further exploration.

It might initially seem surprising that invasive species in their introduced range have larger seeds, but not higher rates of early survival (this is counter to the well-established relationship between seed mass and rates of early seedling survival; Moles and Westoby

2004). However, our result is consistent with a study by Flores-Moreno and Moles

(2013; chapter two of this thesis) that found no significant difference in recruitment success between introduced and native species. One possible explanation is that the potential survival advantage of invasive species in their introduced range is countered

134 by their rapid growth. That is, the observation of invasive species having higher competitive ability under certain conditions but not others (Daehler 2003) could be explained by invasive species having a strategy in which growth rates are maximized

(Grotkopp et al. 2002, Dawson et al. 2011) and biomass increase depends on resource availability (Dawson et al. 2012). However, under this scenario, the trade-off between survival and growth rate (Metcalf et al. 2003, Turnbull et al. 2008, Rose et al. 2009) might be preventing invasive species from achieving higher survival. We need to carefully consider the costs (e.g. higher mortality; lower stem or branches density and therefore less resistant supporting structures) incurred by invasive species having traits related to fast energy absorption, fast growth rates and higher seed production.

The present paper shows that not all fitness-related life history characteristics differ between invasive and non-invasive species, and points to a number of exciting questions to the future. For instance how do invasive species achieve higher seed production?

How lasting are the effects of being in a new range on seed production? What are the costs or trade-offs of invasive species’ life history strategies? We believe that the synthesis of general differences between invasive and non-invasive species ecological strategies at a broad scale is within our reach; clearly this will consolidate decades of efforts put into disentangling invasive species’ ecological strategies.

Many differences in traits associated with fitness between invasive and non-invasive species have been characterized before (Rejmanek and Richardson 1996, Grotkopp and

Rejmánek 2007, Leishman et al. 2007, Moravcová et al. 2010, Davidson et al. 2011,

Dawson et al. 2012). However, we know surprisingly little about the differences in life history characteristics between invasive and non-invasive species that directly impact

135 fitness. We show that overall invasive species do have higher seed production and this is due both to the effect of their nature (species’ intrinsic life history characteristics) and nurture (differences between introduced and native range). Besides seed production, we found no evident connection between invasiveness and dispersal or survival, important life history characteristics affecting the fitness of species. This might be explained by considering the possible trade-off between life history characteristics of invasive species and the myriad of new interactions (biotic and abiotic) that invasive species face in their new range. Our study gives an important step forward by showing that ecological processes in invasive species are complex and will not always lead to high performance.

We hope that future work will expand our knowledge on the life history costs of being invasive. This information will be crucial to understand the ecological strategies of invasive species.

Acknowledgements

H.F.M. was supported by a scholarship from the Evolution & Ecology Research Centre at UNSW. A.T.M. was supported by funding from the Australian Research Council (DP

0984222).

References

Baker, H. G. 1974. The evolution of weeds. Annual Review of Ecology and Systematics

5:1-24.

Biological Records Centre. 2013. Online Atlas of the British and Irish Flora.

Blossey, B. and R. Notzold. 1995. Evolution of increased competitive ability in invasive

nonindigenous plants: A hypothesis. Journal of Ecology 83:887-887.

136 Broennimann, O., U. a. Treier, H. Müller-Schärer, W. Thuiller, a. T. Peterson, and a.

Guisan. 2007. Evidence of climatic niche shift during biological invasion.

Ecology Letters 10:701-709.

Buckley, Y. M., P. Downey, S. V. Fowler, R. Hill, J. Memmot, H. Norambuena, M.

Pitcairn, R. Shaw, A. W. Sheppard, C. Winks, R. Wittenberg, and M. Rees.

2003. Are invasive bigger? A global study of seed size variation in two invasive

shrubs. Ecology 84:1434-1440.

Burns, J. H., E. A. Pardini, M. R. Schutzenhofer, Y. A. Chung, K. J. Seidler, and T. M.

Knight. 2013. Greater sexual reproduction contributes to differences in

demography of invasive plants and their noninvasive relatives. Ecology 94:995-

1004.

Buswell, J. M., A. T. Moles, and S. Hartley. 2011. Is rapid evolution common in

introduced plant species? Journal of Ecology 99:214-224.

Castro-Díez, P., O. Godoy, A. Saldaña, and D. M. Richardson. 2011. Predicting

invasiveness of Australian acacias on the basis of their native climatic affinities,

life history traits and human use. Diversity and Distributions 17:934-945.

Chun, Y. J., M. van Kleunen, and W. Dawson. 2010. The role of enemy release,

tolerance and resistance in plant invasions: linking damage to performance.

Ecology Letters 13:937-946.

CONABIO. 2012. Sistema de información sobre especies invasoras en México.

Comisión Nacional para el Conocimiento y Uso de la Biodiversidad.

Daehler, C. C. 2003. Performance comparisons of co-occurring native and alien

invasive plants: implications for conservation and restoration. Annual Review of

Ecology, Evolution and Systematics 34:183-211.

DAISIE. 2013. Delivering alien invasive species inventories for Europe.

137 Davidson, A. M., M. Jennions, and A. B. Nicotra. 2011. Do invasive species show

higher phenotypic plasticity than native species and, if so, is it adaptive? A

meta-analysis. Ecology Letters 14:419-431.

Dawson, W., M. Fischer, and M. van Kleunen. 2011. The maximum relative growth rate

of common UK plant species is positively associated with their global

invasiveness. Global Ecology and Biogeography 20:299-306.

Dawson, W., R. P. Rohr, M. van Kleunen, and M. Fischer. 2012. Alien plant species

with a wider global distribution are better able to capitalize on increased

resource availability. New Phytologist 194:859-867.

Falster, D. S. and M. Westoby. 2003. Plant height and evolutionary games. Trends in

Ecology & Evolution 18:337-343.

Fitzpatrick, M. C., J. F. Weltzin, N. J. Sanders, and R. R. Dunn. 2007. The

biogeography of prediction error: why does the introduced range of the fire ant

over-predict its native range? Global Ecology and Biogeography 16:24-33.

Flores-Moreno, H. and A. T. Moles. 2013. A comparison of the recruitment success of

introduced and native species under natural conditions. PLoS ONE 8:e72509.

Flores-Moreno, H., F. J. Thomson, D. I. Warton, and A. T. Moles. 2013. Are introduced

species better dispersers than native species? A broad comparative study of seed

dispersal distance. PLoS ONE 8:e68541.

Gallagher, R. V., L. J. Beaumont, L. Hughes, and M. R. Leishman. 2010. Evidence for

climatic niche and biome shifts between native and novel ranges in plant species

introduced to Australia. Journal of Ecology 98:790-799.

Gallagher, R. V., M. R. Leishman, J. T. Miller, C. Hui, D. M. Richardson, J. Suda, and

P. Trávníček. 2011. Invasiveness in introduced Australian Acacias: the role of

species traits and genome size. Diversity and Distributions 17:884-897.

138 Grotkopp, E. and M. Rejmánek. 2007. High seedling relative growth rate and specific

leaf area are traits of invasive species: Philogenetically independent contrasts of

woody angiosperms. American Journal of Botany 94:526-532.

Grotkopp, E., M. Rejmánek, and T. L. Rost. 2002. Toward a causal explanation of plant

invasiveness: seedling growth and life-history strategies of 29 Pine (Pinus)

species. American Naturalist 159:396-419.

Hawkes, C. V. 2007. Are invaders moving targets? The generality and persistence of

advantages in size, reproduction, and enemy release in invasive plant species

with time since introduction. American Naturalist 170:832-843.

Jongejans, E., K. Shea, O. Skarpaas, D. Kelly, A. W. Sheppard, and T. L. Woodburn.

2008. Dispersal and demography contributions to population spread of Carduus

nutans in its native and invaded ranges. Journal of Ecology 96:687-697.

Knapp, S. and I. Kühn. 2012. Origin matters: widely distributed native and non-native

species benefit from different functional traits. Ecology Letters 15:696-703.

Kolar, C. S. and D. M. Lodge. 2001. Progress in invasion biology: predicting invaders.

Trends in Ecology & Evolution 16:199-204.

Küster, E. C., W. Durka, I. Kühn, and S. Klotz. 2009. Differences in the trait

compositions of non-indigenous and native plants across Germany. Biological

Invasions 12:2001-2012.

Leishman, M. R., T. Haslehurst, A. Ares, and Z. Baruch. 2007. Leaf trait relationships

of native and invasive plants: community- and global-scale comparisons. New

Phytologist 176:635-643.

Liu, H. and P. Stiling. 2006. Testing the enemy release hypothesis: a review and meta-

analysis. Biological Invasions 8:1535-1545.

139 Mason, R. A. B., J. Cooke, A. T. Moles, and M. R. Leishman. 2008. Reproductive

output of invasive versus native plants. Global Ecology and Biogeography

17:633-640.

Metcalf, J. C., K. E. Rose, and M. Rees. 2003. Evolutionary demography of monocarpic

perennials. Trends in Ecology & Evolution 18:471-480.

Moles, A. T., D. D. Ackerly, C. O. Webb, J. C. Tweddle, J. B. Dickie, A. J. Pitman, and

M. Westoby. 2005. Factors that shape seed mass evolution. Proceedings of the

National Academy of Sciences of the United States of America 102:10540-

10544.

Moles, A. T., D. S. Falster, M. R. Leishman, and M. Westoby. 2004. Small-seeded

species produce more seeds per square metre of canopy per year , but not per

individual per lifetime. Journal of Ecology 92:384-396.

Moles, A. T., D. I. Warton, L. Warman, N. G. Swenson, S. W. Laffan, A. E. Zanne, A.

Pitman, F. a. Hemmings, and M. R. Leishman. 2009. Global patterns in plant

height. Journal of Ecology 97:923-932.

Moles, A. T. and M. Westoby. 2004. Seedling survival and seed size: a synthesis of the

literature. Journal of Ecology 92:372-383.

Moore, L. B. and E. Edgar. 1970. Flora of New Zealand First electronic edition

Landcare Research June 2004. Transcr. A.D. Wilton and I.M.L. Andres.

Morales, C. L. and A. Traveset. 2009. A meta-analysis of impacts of alien vs. native

plants on pollinator visitation and reproductive success of co-flowering native

plants. Ecology Letters 12:716-728.

Moravcová, L., P. Pyšek, V. Jarosik, V. Havlícková, and P. Zákravský. 2010.

Reproductive characteristics of neophytes in the Czech Republic: traits of

invasive and non-invasive species. Preslia 82:365-390.

140 NRCS. 2013. The PLANTS Database. National Plant Data Team, Greensboro, NC

27401-4901 USA.

Parker, J. D., M. E. Torchin, R. A. Hufbauer, N. P. Lemoine, C. Alba, D. M.

Blumenthal, O. Bossdorf, J. E. Byers, A. M. Dunn, R. W. Heckman, M. Hejda,

V. Jarošík, A. R. Kanarek, L. B. Martin, S. E. Perkins, P. Pyšek, K. A.

Schierenbeck, C. Schloder, R. D. Van Klinken, K. J. Vaughn, W. Williams, and

L. M. Wolfe. 2013. Do invasive species perform better in their new ranges?

Ecology 94:985-994.

Petitpierre, B., C. Kueffer, O. Broennimann, C. Randin, C. Daehler, and A. Guisan.

2012. Climatic niche shifts are rare among terrestrial plant invaders. Science

(New York, N.Y.) 335:1344-1348.

Pyšek, P., J. Daniehlka, J. Sadlo, J. Chrtek Jr, M. Chytry, V. Jarošík, Z. Kaplan, F.

Krahulec, L. Moravcová, J. Pergl, K. Stajerova, and L. Tichy. 2012. Catalogue

of alien plants of the Czech Republic (2nd edition): checklist update , taxonomic

diversity and invasion patterns. Preslia 84:155-255.

Pyšek, P., V. Jarošík, J. Pergl, R. Randall, M. Chytry, I. Kühn, L. Tichy, J. Danihelka, J.

Chrtek jun, and J. Sádlo. 2009. The global invasion success of Central European

plants is related to distribution characteristics in their native range and species

traits. Diversity and Distributions 15:891-903.

Pyšek, P. and D. M. Richardson. 2007. Traits associated with invasiveness in alien

plants: where do we stand? Pages 97-125 in W. Nentwig, editor. Biological

Invasions, Ecological Studies Vol. 193. Springer-Verlag, Berlin.

R Core Team. 2013. R: a language and environment for statistical computing.in R

Foundation for Statistical Computing, editor., Vienna, Austria.

141 Ramula, S., T. M. Knight, J. H. Burns, and Y. M. Buckley. 2008. General guidelines for

invasive plant management based on comparative demography of invasive and

native plant populations. Journal of Applied Ecology 45:1124-1133.

Randall, R. P. 2007a. Global Compedium of Weeds. RG and FJ Richardson,

Melbourne, Australia.

Randall, R. P. 2007b. The introduced and its weed status. CRC for Au

edition. CRC weed management, Adelaide, Australia.

Rayamajhi, M. B., T. K. Van, J. A. Goolsby, P. D. Pratt, and A. Racelis. 2002.

Biological Attributes of The Canopy-held Melaleuca Seeds in Australia and

Florida, U.S. Journal of Aquatic Plant Management 40:87-91.

Rejmánek, M. 1996. A theory of seed plant invasiveness: the first sketch. Biological

Conservation 78:171-181.

Rejmanek, M. and D. M. Richardson. 1996. What attributes make some plant species

more invasive? Ecology 78:171-181.

Richardson, D. M. and P. Pyšek. 2012. Naturalization of introduced plants: ecological

drivers of biogeographical patterns. New Phytologist 196:383-396.

Richardson, D. M., P. Pyšek, M. Rejmánek, M. G. Barbour, F. Dane, and C. J. West.

2000. Naturalization and invasion of alien plants: concepts and definitions.

Diversity and Distributions 6:93-107.

Rose, K. E., R. L. Atkinson, L. a. Turnbull, and M. Rees. 2009. The costs and benefits

of fast living. Ecology Letters 12:1379-1384.

Schlaepfer, D. R., M. Glättli, M. Fischer, and M. van Kleunen. 2010. A multi-species

experiment in their native range indicates pre-adaptation of invasive alien plant

species. New Phytologist 185:1087-1099.

The Council of Heads of Australasian Herbaria. 1999. Australia's Virtual Herbarium

142 Thomson, F. J., A. T. Moles, T. D. Auld, and R. T. Kingsford. 2011. Seed dispersal

distance is more strongly correlated with plant height than with seed mass.

Journal of Ecology 99:1299-1307.

Turnbull, L. a., C. Paul-Victor, B. Schmid, and D. W. Purves. 2008. Growth rates, seed

size, and physiology: do small-seeded species really grow faster? Ecology

89:1352-1363.

USDA ARS National Genetic Resources Program. 2013. Germplasm Resources

Information Network - (GRIN) [Online Database]. National Germplasm

Resources Laboratory, Beltsville, Maryland. van Kleunen, M. and S. D. Johnson. 2007a. Effects of Self-Compatibility on the

Distribution Range of Invasive European Plants in North America. Conservation

Biology 21:1537-1544. van Kleunen, M. and S. D. Johnson. 2007b. South African Iridaceae with rapid and

profuse seedling emergence are more likely to become naturalized in other

regions. Journal of Ecology 95:674-681. van Kleunen, M., S. D. Johnson, and M. Fischer. 2007. Predicting naturalization of

southern African Iridaceae in other regions. Journal of Applied Ecology 44:594-

603. van Kleunen, M., E. Weber, and M. Fischer. 2010. A meta-analysis of trait differences

between invasive and non-invasive plant species. Ecology Letters 13:235-245.

Warton, D. I. and F. K. C. Hui. 2011. The arcsine is asine: the analysis of proportions in

ecology. Ecology 92:3-10.

Westoby, M., D. S. Falster, A. T. Moles, P. A. Vesk, and I. J. Wright. 2002. Plant

ecological strategies: some leading dimensions of variation between species.

Annual Review of Ecology and Systematics 33:125-159.

143

Supporting information

Supporting table 1. Relationship between plant height and canopy area.

Supporting table 2. Simple comparison of life history characteristic and traits between species with different level of invasiveness.

Supporting table 3. Comparison of life history characteristics of species with different level of invasiveness once the effect of ecological relevant traits had been accounted for.

Supporting table 4. Comparison of life history characteristic and traits between native species with different level of invasiveness.

Supporting table 5. Comparison of life history characteristic and traits between invasive species in their native range and invasive species in their introduced range.

144 Nature vs. nurture

Supporting table 1. Relationship between plant height and canopy area.

Terms Estimate Standard error P

Intercept -0.78 0.10 < 0.0001

Plant height 2.40 0.15 < 0.0001

145 Nature vs. nurture

Supporting table 2. Simple comparison of life history characteristic and traits between species with different level of invasiveness.

All analyses were ran on logit-transform survival data and log10-transform seed mass, plant height, seed production and seed dispersal data.

1) Analysis for seed mass Sum of Terms d.f. P squares Species’ level of 94.29 2 <0.0001 invasiveness Residuals 1641.13 1068 2) Analysis for plant height Sum of Terms d.f. P squares Species’ level of 4.1 2 0.006 invasiveness Residuals 492.54 1225 3) Analysis for seed production (seeds.individual-1.yr -1) Sum of Terms d.f. P squares Species’ level of 29.7 2 <0.0001 invasiveness Residuals 434.4 402 4) Anaylsis for mean dispersal distance Sum of Terms d.f. P squares Species’ level of 4.14 2 0.2 invasiveness Residuals 301.13 235 5) Analysis for maximum dispersal distance Sum of Terms d.f. P squares Species’ level of 0.94 2 0.62 invasiveness Residuals 233.81 240 6) Analysis for survival through germination Sum of Terms d.f. P squares

146 Nature vs. nurture

Species’ level of 8.48 2 0.32 invasiveness Residuals 946.23 253 7) Analysis for early seedling survival Sum of Terms d.f. P squares Species’ level of 5.78 2 0.4 invasiveness Residuals 381.96 119

147 Nature vs. nurture

Supporting table 3. Comparison of life history characteristics of species with different level of invasiveness once the effect of ecological relevant traits had been accounted for.

We ran linear models to test whether seed production, seed dispersal and survival of species with different levels of invasiveness differ once ecological important traits (plant height and seed mass) have been accounted for. For the seed production comparison our predictor variables were plant height, seed mass, level of invasiveness and their interactions and our dependent variable was seed production per individual per year. Then, for the seed dispersal comparison our predictor variables were plant height, seed mass, level of invasiveness and their interactions and our response variables were mean or maximum dispersal distance.

Finally, for our early seedling survival comparison our predictor variables were seed mass and level of invasiveness and the dependent variable was early seedling survival.

All analyses were ran on logit-transform survival data and log10-transform seed mass, plant height, seed production and seed dispersal data.

1) Analysis for seed production (seeds.individual-1.yr-1) Terms Sum of squares d.f. P Level of invasiveness 16.82 2 0.0001 < Seed mass 46.18 1 0.0001 < Plant height 47.05 1 0.0001 Level of invasiveness × Seed mass 1.2 2 0.52 Level of invasiveness × Plant height 3.05 2 0.19 Residuals 359.7 396

148 Nature vs. nurture

2) Analysis for mean dispersal distance Terms Sum of squares d.f. P Level of invasiveness 0.83 2 0.62 Seed mass 5.76 1 0.01 < Plant height 46.89 1 0.0001 Level of invasiveness × Seed mass 2.97 2 0.18 Level of invasiveness × Plant height 2.31 2 0.26 Residuals 196.47 229 3) Analysis for maximum dispersal distance Terms Sum of squares d.f. P Level of invasiveness 2.29 2 0.26 Seed mass 0.99 1 0.28 < Plant height 14.74 1 0.0001 Level of invasiveness × Seed mass 2.96 2 0.18 Level of invasiveness × Plant height 4.81 2 0.06 Residuals 200.45 234 4) Analysis for early seedling survival Terms Sum of squares d.f. P Level of invasiveness 2.21 2 0.69 Seed mass 20.72 1 0.01 Level of invasiveness × Seed mass 6.49 2 0.35 Residuals 354.75 116

149 Nature vs. nurture

Supporting table 4. Comparison of life history characteristic and traits between native species with different level of invasiveness.

All analyses were ran on logit-transform survival data and log10-transform seed mass, plant height, seed production and seed dispersal data.

1) Analysis for seed mass Terms Sum of squares d.f. P Level of invasiveness 119.19 2 < 0.0001 Residuals 1453.77 944 2) Analysis for plant height Terms Sum of squares d.f. P Level of invasiveness 4.17 2 0.006 Residuals 449.29 1085 3) Analysis for seed production (seeds.individual-1 .yr-1) Terms Sum of squares d.f. P Level of invasiveness 8.2 2 0.02 Residuals 346.9 350 4) Analysis for mean dispersal distance Terms Sum of squares d.f. P Level of invasiveness 6.009 2 0.1 Residuals 252.1 195 5) Analysis for maximum dispersal distance Terms Sum of squares d.f. P Level of invasiveness 0.062 2 0.96 Residuals 162 191 6) Analysis for survival through germination Terms Sum of squares d.f. P Level of invasiveness 5.21 2 0.5 Residuals 750.03 200 7) Analysis for early seedling survival Terms Sum of squares d.f. P Level of invasiveness 6.19 2 0.24 Residuals 169.46 79

150 Nature vs. nurture

Supporting table 5. Comparison of life history characteristic and traits between invasive species in their native range and invasive species in their introduced range.

All analyses were ran on logit-transform survival data and log10-transform seed mass, plant height, seed production and seed dispersal data.

1) Analysis for seed mass Terms Sum of squares d.f. P Level of invasiveness 18.84 1 < 0.0001 Residuals 510.49 440 2) Analysis for plant height Terms Sum of squares d.f. P Level of invasiveness 0.215 1 0.44 Residuals 167.724 457 3) Analysis for seed production (seeds.individual-1.yr-1) Terms Sum of squares d.f. P Level of invasiveness 10.45 1 0.004 Residuals 138.62 112 4) Analysis for yearly seed production per indivudla per year (seed mass × seeds.individual-1.yr-1) Terms Sum of squares d.f. P Level of invasiveness 142.91 1 < 0.001 Residuals 1015.04 112 5) Analysis for mean seed dispersal distance Terms Sum of squares d.f. P Level of invasiveness 0.664 1 0.47 Residuals 90.735 70 6) Analysis for maximum seed dispersal distance Terms Sum of squares d.f. P Level of invasiveness 0.45 1 0.53 Residuals 95.52 85 7) Analysis for survival through germination Terms Sum of squares d.f. P Level of invasiveness 1.96 1 0.48 Residuals 356.33 92

151 Nature vs. nurture

8) Analysis for early seedling survival Terms Sum of squares d.f. P Level of invasiveness 1.86 1 0.53 Residuals 265.15 58

152 Phenotypic change in invasive species

Chapter four

In the beginning: Phenotypic change in three invasive

species through their first 200 years since introduction.

Habacuc Flores-Moreno, Edgar S. García-Treviño, Andrew D. Letten & Angela T.

Moles

The study was conceived and designed by HFM and ATM. HFM, EGT and ADL

performed the experiments.HFM collected and collated the data. HFM, ADL and EGT

analysed the data. EGT provided analytical tools. The manuscript writing was led by

HFM, but ATM and EGT contributed substantially to it.

Abstract

Previous studies have demonstrated that most introduced species often go through rapid phenotypic change during their first decades to centuries of being introduced to a new range. However, little is known about the trends these phenotypic changes follow through time. Using herbarium specimens we track changes in the leaf area, leaf shape and plant height of three species ( ciliatum, Senecio squalidus and Veronica

153 Phenotypic change in invasive species

persica) through their first ~200 years since introduction to the United Kingdom (U.K.).

All three species showed fluctuating direction and strength of phenotypic change through time. None of our species showed a lag phase in the phenotypic change at the beginning of the invasion. Next, we asked whether phenotypic changes were ongoing, or whether the species had reached a new equilibrium state. All three species were still changing in at least one trait hundreds of years after their introduction. This suggests that some introduced species are yet to demonstrate their full potential as invaders.

Introduction

Although not all, most introduced species undergo behavioural, phenological and morphological changes when confronted with novel conditions (Orr and Smith, 1998;

Thompson, 1998; Lambrinos, 2004; Buswell et al., 2011). However, we know remarkably little about the trajectories that introduced species follow through time

(Bone and Farres, 2001; Westley, 2011). This situation limits our understanding about the process of invasion (and thereby our power to predict the establishment and ecological effects of invasive species), and our understanding of how species respond to environmental changes. The main aim of this paper is to fill this knowledge gap, by quantifying phenotypic change in three plant species though their first 200 years since introduction.

Theory suggests that the most rapid changes in introduced species will occur very soon after their introduction to a new range. This is because it is then that the strongest selective challenges are faced (Thompson, 1998; Reznick and Ghalambor, 2001;

Lambrinos, 2004; Shine, 2012), and because high population growth rates at the

154 Phenotypic change in invasive species

beginning of the invasion could enhance the opportunity for rapid evolution (Reznick and Ghalambor, 2001; Vellend et al., 2007). Conversely, other studies suggest that the early low densities and/or poor adaptation of introduced species to their new ranges might result in slow rates of adaptation in the early stages of introduction (Mooney and

Cleland, 2001; Sakai et al., 2001; Holt et al., 2005; Suarez and Tsutsui, 2008). Previous studies have mostly described introduced species’ lags in population growth rates or range expansion (Pyšek and Prach, 1993; Kowarik, 1995; Williamson et al., 2005;

Daehler, 2009; Aikio et al., 2010). These studies often suggest that lag phases in population growth or spread could be explained by a lag in species’ ability to adapt their phenotype to new conditions (Hobbs and Humphriest, 1995; Sakai et al., 2001; Crooks,

2005; Holt et al., 2005). However, no previous study has offered an empirical test for a lag phase in phenotypic change in introduced species. Our first aim is to determine whether introduced species experience a lag phase in their phenotypic change.

Determining the trajectories of phenotypic change in invasive species is crucial for understanding how ecological traits are influenced by new ranges. This ultimately impacts our knowledge of how introduction events could affect the transition between levels of invasiveness (pathways of invasion), and in general, how plant populations adapt when exposed to new environmental conditions.

Determining whether introduced species keep changing long after being introduced is important to fully understand the long term dynamics of species introductions to new ecosystems. If introduced species are no longer changing, it is likely that they have reached a period of stabilizing selection. However if they are still changing years after being introduced, and thus still becoming better adapted to their new environment, then

155 Phenotypic change in invasive species

it is probable that we are yet to see introduced species’ full potential as invaders.

Evolutionary theory predicts that after a period of rapid adaptation to their new biotic and abiotic conditions, introduced species will experience different selective pressures that would change the trajectory of phenotypic change from directional to fluctuating around a new optimum value, stabilizing the phenotypic change (Thompson, 1998).

Previous studies have normally compared absolute rates of change between native and introduced populations (Blossey and Notzold, 1995; Daehler and Strong, 1997; Huey,

2000) or between two points in time (Carroll et al., 1997; Phillips et al., 2006), or have assumed the change to be constant throughout the invasion due to the lack of sufficient time series data (Buswell et al., 2011). No one has ever shown whether introduced species keep changing long after being introduced to a new range or whether they reach a new equilibrium. This is our second aim.

Native species could be undergoing rapid phenotypic change just as much as our invasive species. This is because, like invasive species, native species go through changes as a response to current environmental conditions. For instance, previous studies have shown that native species are undergoing rapid phenotypic change in response to global climate change (Jump and Penuelas, 2005; Hoffmann and Sgrò,

2011). Other studies showed that native species undergo morphological, phenotypic, and physiological changes in response to new interactions with invasive predators, competitors and parasites (Strauss, 2006). A compilation by Westley (2011) suggests that introduced and native species do not differ in their rates of phenotypic change.

However, studies were pairs of invasive and native species have been compared showed that invasive species tend to have higher rates of change and tend to change more often

156 Phenotypic change in invasive species

than native species. For instance, Buswell et al. (2011) demonstrated that 16 out of 23 invasive species compared to one out of eighth native species in Australia showed phenotypic change through time. Similarly, vote-counting and meta-analytical studies showed that introduced populations tend to go through changes in defence-, reproductive-, size-, and growth-related traits (Bossdorf et al., 2005; Felker-Quinn et al., 2013; Parker et al., 2013). Native and invasive species in our study could both be undergoing rapid phenotypic change. This would affect the interpretation of our results.

For instance, if both native and invasive species are changing, this would suggest that the changes are related to a selective pressure common to both native and invasive species (e.g. climate or habitat change), rather than to invasive species’ response to the introduction event. Our third aim was to assess whether three native species with similar habit, habitat and overlapping distribution as our invasive species were experiencing similar rates of phenotypic change.

In summary, the main aim of this paper is to quantify the trajectory of the phenotypic change on three introduced species. We will address three hypotheses:

1) Introduced species will go through a lag phase in their phenotypic change at the

beginning of the invasion.

2) Introduced species’ phenotypes are no longer changing long after their

introduction to a new range.

3) Introduced species are changing more than native species for the same period of

time.

157 Phenotypic change in invasive species

Methods

We studied three annual herbs introduced to the United Kingdom (U.K.) : Senecio squalidus L. (Asteraceae, first recorded in the wild in the U.K 1794), Epilobium ciliatum Raf. (, first collected in the U.K. in 1891) and Veronica persica

Poir. (Plantaginaceae, first recorded in the wild in 1826; Biological Records Centre,

2013). Our selection of species was based on the fact that their herbarium collections span from the same year or decade they were introduced to the U.K. to the present day.

For S. squalidus the herbarium specimens range from 1792-2012, for V. persica from

1826-2009 and for E. ciliatum from 1891-2004. The oldest specimen for Senecio squalidus is held at Cambridge University; while the collection predates the “Atlas of the British and Irish flora” (Biological Records Centre, 2013) account of first record in the wild, it has been properly verified. To determine whether native species were changing less than introduced species, we sampled three native species with the same habit, habitat and overlapping distribution as the invasive species. The native species that we selected were Epilobium montanum, Senecio squalidus and Stellaria media.

Specimens are held in the Natural History Museum, Royal Kew Botanical Gardens,

University of Oxford, Manchester Museum, Liverpool Museum and University of

Sheffield herbaria. In total we sampled 1224 herbarium specimens, 518 and 706 specimens from three different invasive and three different native species respectively.

We measured all available specimens at these herbaria.

Herbarium methods

We began by measuring the mean plant height, leaf area and leaf shape following the methods of Buswell et al. (2011) and Dalrymple et al. (unpublished).

158 Phenotypic change in invasive species

We measured plant height as the shortest distance between the highest vegetative tissue of the plant and ground level for the Epilobium and Senecio species. For the prostrate species (Veronica persica and Stellaria media) we measured plant height as the stretched distance of a stem from its youngest fully active expanded apical leaf to the ground level (stretched length; Cornelissen et al., 2003). Plant height data were collected for every herbarium specimen for which we could clearly identify where the roots began (following Buswell et al., 2011).

We measured the length and width of three to four fully expanded of every herbarium specimen for which plant height had been measured. We measured leaf length as the longest distance between the base of the leaf (without considering the petiole) and the tip of the leaf (Buswell et al., 2011). We considered leaf width as the maximum diameter of the largest imaginary circle that could fit on the leaf (Westoby,

1998; Cornelissen et al., 2003; Buswell et al., 2011).

We calculated leaf shape as the ratio between leaf width and leaf length (Buswell et al.,

2011).

We considered leaf area as the area of the upper surface of the leaf (Cornelissen et al.,

2003). First, we quantified the leaf area, leaf width and leaf length of three to four fully expanded fully grown leaves of 10 to 15 random specimens per species using Image J

(Rasband, 1997-2013). Then, using a linear regression we quantified the relationship between the product of leaf width by length and leaf area (R2 > 0.90 for each species).

159 Phenotypic change in invasive species

Finally, we calculated the leaf area of all the specimens as a product of leaf length, width and a shape parameter following Buswell et al. (2011).

Data analyses

We collected time series data for leaf area, leaf shape and plant height for each of our three species. However, for simplicity we will refer to these data as the collective time series from now on. The time series for each trait included in this study spanned more than 30 years, and data points ≥ 30 years away from the next closest data point were excluded (Hendry and Kinnison, 1999; Buswell et al., 2011). In all analyses, we weighted the data in inverse proportion to the number of individuals on each herbarium sheet so that all sheets had a statistical weight equal to one (Buswell et al., 2011). We did this to account for the fact that individuals coming from the same herbarium sheet might be more similar because shared environmental conditions or they may even share same predecessors (Buswell et al., 2011). All morphological measures were log10- transformed prior to analysis.

To describe the trajectory of change that our introduced species followed the first 200 years after introduction we used the Discrete Wavelet Transformation (DWT; Mallat,

2008). The DWT is a multi-resolution decomposition transform that allows the user to separate different scales of information by projecting the data onto a set of basis functions (Mallat, 2008). The DWT allows analysis of the data into detailed coefficients

(short term trends or fine scale information) and approximation coefficients (long term trends or large-scale information; Mallat, 2008). Wavelet transformations are a powerful tool in the analyses of time series because they do not assume stationarity (i.e. they do

160 Phenotypic change in invasive species

not assume that the statistical properties of the time series do not vary with time;

Cazelles et al., 2008; Mallat, 2008). We used the DWT method to calculate a global phenotypic change for each species and to calculate phenotypic change coefficients for each trait. For details on the DWT method see supporting information 1.

We ran linear regressions to determine the rate of change on introduced species phenotype across the whole period of observation (following Buswell et al., 2011). In these linear models the response variables were the species’ traits (leaf shape, leaf area or plant height) and the predictor variable was time.

To determine whether introduced species go through a lag phase at the beginning of the invasion we fit a linear model and a non-linear power model to the first 100 years of data for our three species. In these models the predictor variable was time (years) and the response variable were leaf area, leaf shape or plant height. Then using an ANOVA test we compare whether the linear or non linear power model explained more variation in the change in traits values through time. If a power model shows a significantly better fit than a linear model, then this would suggest that the rate of change is initially slow picking up speed through time.

To determine whether introduced species are still undergoing phenotypic change long after their introduction, we ran a linear model using the trait data for the last 50 years where our predictor variable was time (years) and our response variable was leaf area, leaf shape or plant height. The analyses of the lag phase in the rate of change and of the

161 Phenotypic change in invasive species

undergoing phenotypic change long after introduction were done using R 2.15.3 (R

Core Team, 2013).

Both ecological and evolutionary mechanisms could be responsible for the phenotypic change trajectories observed in our species and we cannot rule out the possibility that the changes and patterns we observed are due to phenotypic plasticity or have little impact on the genotypic variation of our species.

Data considerations

1. Geographic location

As introduced plants spread out through geographic space, they might encounter different environmental conditions. Thus phenotypic change through time can be confounded with phenotypic change due to differences in environmental conditions (see

Buswell et al., 2011). Therefore, we asked whether plants from different regions had different rates of morphological change through time (Buswell et al., 2011). We divided the U.K. into three major geographical regions in a north to south gradient (Boreo- temperate, Temperate and Southern-temperate; Preston and Hill, 1997). These regions reflect a latitudinal gradient in temperature that is a major driver of floristic differences in the U.K (Preston and Hill, 1997). For our three species we excluded data from the

Boreo-temperate zone because we had low sample sizes in this zone (n ≤ 2 for all the species). Then we ran an ANCOVA to test whether plants from the Temperate and

Southern-temperate regions had differences in the magnitude and direction of change in their phenotypic traits. Our predictor variables were time and region, our response

162 Phenotypic change in invasive species

variable was change in leaf area, leaf shape or plant height. We found no significant difference between regions for any of our species (P > 0.05 for all species and regions; supporting table 1). Therefore all the trait data in this study were analysed without divisions within species for the Temperate and Southern-temperate regions. The low incidence of data for our species from the Boreo-temperate region coincides with the fact that our three species were first introduced to the Temperate or Southern-temperate regions in the U.K (Biological Records Centre, 2013).

2. Native control

In order to make a fair comparison we only included native species’ data from the same regions and years as the invasive species’ data. That is, we only included native species’ data from the Temperate and Southern-temperate regions as only data from these regions was included for the invasive species. We also only included native species’ data from approximately the same range of years as the data for invasive species. Then, we ran ANCOVA tests to assess whether native plants from different regions showed difference in the magnitude and direction of change in their phenotypic traits through time (Supporting table 1). We only found a significant interaction between region and year for Senecio vulgaris’ height (P = 0.03). Thus, for Senecio vulgaris we fitted linear models separately for the Temperate and Southern-temperate regions. Then, to assess whether native species where going through phenotypic change through time we ran a linear model, where the predictor variable was year and the response variable were the species’ traits. To determine whether native and invasive species differ in their rates of phenotypic change we ran ANCOVA tests. We only compared the rate of phenotypic change of native and invasive species when we found a significant change in the rate of

163 Phenotypic change in invasive species

phenotypic change through time for native species. In these ANCOVAs the predictor variables were species’ status (invasive or native), time in years, and their interaction and the response variables were the species’ traits (leaf shape, leaf area or plant height).

Then we ran a linear model to test whether the phenotypic changes on native species were significant

164 Phenotypic change in invasive species

Results

Phenotypic changes have fluctuated in direction and strength across time for all our species through the first 200 years after their introduction (Figure 1). On average,

Epilobium ciliatum showed a significant 50% decrease in leaf area (P = 0.003, Figure 1 a), and no significant changes in leaf shape and plant height (both P = 0.16, Figure 1 b and c respectively). In the case of Senecio squalidus we saw a significant 50% decrease in leaf area (P = 0.02, Figure 1 d) and 19% significant increase in plant height (P =

0.003, Figure 1 f). Leaf shape did not showed any significant change (P = 0.4, Figure 1 e). For Veronica persica leaf area, leaf shape and plant height showed significant changes through time (P ≤ 0.03, Figure 1 g, h and i respectively). Leaf area showed a

26% increase, leaf shape showed a 14% decrease (leaves became more narrow), while plant height showed a 5% increase.

165 Phenotypic change in invasive species

Figure 1. Change in leaf area (triangles), leaf shape (circles) and plant height (squares) through time in Epilobium cilaitum (a to c), Senecio squalidus (d to e) and Veronica persica (g to i). P-values are given when we found significant changes through time.

The direction and the strength of the phenotypic changes fluctuate in direction and strength across time for all of our species. None of our species registered phenotypic change in a single burst, but rather the change in the species’ traits seemed to happen in sprouts of change across time (Figure 2). The maximum global phenotypic change

((E(t); Figure 2) observed in Epilobium ciliatum was 38 years after introduction (ca.

1930s; Figure 2 a), in the case of Senecio squalidus was observed 113 years after introduction (ca. 1930s; Figure 2 b) and for Veronica persica 178 years after

166 Phenotypic change in invasive species

introduction (2000s; Figure 2 c). For details on the maximum rates of change for each trait see supporting information 2.

Figure 2. Rates of phenotypic change on leaf shape (Eleaf shape (t)), leaf area (Etrait(t)), plant height, and global phenotypic change (E(t)) through time for Epilobium ciliatum,

Senecio squalidus, and Veronica persica. Time is in years. Each graph is subdivided into leaf shape, leaf area and plant height. The rates are calculated as absolute values of change. Absolute rates of change are calculated on normalized (mean = 0 and unit standard deviation) trait values.

Is there any evidence for a lag phase?

When we asked whether introduced species had a lag phase we found that the linear models explained significantly more variation than did the power models (P < 0.05)

(Supporting table 2). For Senecio squalidus’ leaf shape the linear model had a marginally significant better fit than the power model (P = 0.056; Supporting table 2).

167 Phenotypic change in invasive species

That is, introduced species do not seem to go through a lag phase. When we increased the time frame to encompass the complete period of observation for each species we found qualitatively similar results. In this case for all the species the linear models explained significantly more variation than did the power models (P < 0.05).

Are the species still changing?

When we compared the rates of change in the traits of our introduced species for the last

50 years (~1950s- 2000s) we found that each of our species showed significant changes in at least one trait. Epilobium ciliatum and Veronica persica showed a 68% and 74% significant decrease in leaf area (P = 0.04 and 0.01, respectively). Senecio squalidus showed a significant 65% increase in leaf shape (P = 0.04). No other trait for any of our species registered significant changes (P > 0.13) in the last 50 years.

Native species control

Of our native species, only Stellaria media and Senecio vulgaris from the temperate region showed significant change through time in only one trait (P = 0.0005). The native Stellaria media became 12 % more narrow, while the native Senecio vulgaris from the temperate region showed a 13 % increase in height. When we compared these two species with their respective invasive pair we found that the native Stellaria media and the invasive Veronica persica had a significant difference in magnitude and direction of change in leaf shape (P < 0.0001). The invasive Veronica persica showed a significant 57 % increase in leaf shape, while the native Stellaria media showed an increase a five times lower increase in leaf shape (12 %). We did not find a significant

168 Phenotypic change in invasive species

difference in the direction or magnitude of change between the native Senecio vulgaris from the Temperate region and the invasive Senecio squalidus (P ≥ 0.59). However the invasive Senecio squalidus’ change in height through time was 1.40 times faster than that of the native Senecio vulgaris from the temperate region.

Discussion

We have provided some crucial contributions to the little that is known about the long term patterns of phenotypic changes in introduced species. Our most important findings were: 1) No evidence of a lag phase, 2) introduced species are still changing long after being introduced in their new range, 3) although there are overall significant changes in these species through time, the changes are not in a consistent progression, but rather fluctuate through time.

Although there is evidence for lag phases in population size and range expansion of introduced species (summarized in Crooks, 2005), our result is contrary to the proposition that lags in the spread or population increase of introduced species results in lags in phenotypic change (Hobbs and Humphriest, 1995; Sakai et al., 2001; Crooks,

2005; Holt et al., 2005; Facon et al., 2006). While low population sizes in combination with low genetic variation (Mooney and Cleland, 2001; Lee, 2002; but see Vellend et al., 2007) at the beginning of the invasion might delay the potential for adaptive changes, it is then when the directional selection should be at its strongest. That is, there might not be a lag phase in an introduced species’ phenotypic change because the difference between an introduced species’ expressed phenotype and the optimal phenotype for the new environment (the phenotypic mismatch) should be greatest at the beginning of the invasion. This is both encouraging when we think about native species

169 Phenotypic change in invasive species

facing new environmental challenges, and discouraging as this might allow introduced species to colonize diverse environments.

Our result also gives us insight into the mechanisms behind species’ adaptation to novel environmental conditions. It is also encouraging as it suggests that species do not necessarily show a slow response in the face of environmental changes. Overall, our results add to a growing body of evidence showing that species (including animals) are much more evolutionarily fluid than we have given them credit for (see Williams and

Moore, 1989; Carroll et al., 1997; Phillips et al., 2006; Hendry et al., 2007; Zangerl et al., 2008; Westley, 2011).

Three out of three species showed evidence of change at least in one trait in the last fifty years (~1950s- 2000s), Epilobium ciliatum and Veronica persica showed a decrease in leaf area and Senecio squalidus showed a increase in leaf shape. That is, we have shown that introduced species have the capacity to keep changing many years after being introduced to a new range. These results suggest that introduced species are still becoming better adapted to the local conditions long after their establishment, and this has two major implications. First, it is possible that as introduced species become more adapted to local conditions they might become better invaders. Second, species’ introductions could lead to evolutionary diversifications (Vellend et al., 2007). The capacity to keep changing long after being introduced could allow introduced species to spread to more and more diverse environments, leading to novel species interactions. It also indicates they could be ecologically flexible in the face of climate or ecosystem change. Evolutionary diversification in such scenarios seems likely.

170 Phenotypic change in invasive species

Phenotypic changes often vary in intensity and direction through time (Hendry and

Kinnison, 1999). Our data suggest that introduced species might never reach equilibrium, with traits always fluctuating around an optimum or average trait value.

Rapid phenotypic changes are common across introduced species (Buswell et al., 2011) and shortly after their introduction introduced species experience directional selection to some extent (Reznick and Ghalambor, 2001; Lambrinos, 2004; Westley, 2011).

However, increases in abundance and/or longer residence time could lead to exposures to new negative and positive interactions that could reshape the strength and/or direction of the rates of phenotypic change on introduced species (Thompson, 1998).

For instance, invasive species might benefit from enemy release early on in after being introduced to their new range (Blossey and Notzold, 1995; Keane and Crawley, 2002) partially because of their low abundance. However as they become more abundant they may experience more herbivory, because more abundant or widespread plants experience more herbivory than do less widespread or rarer species (Strong et al.,

1984). If introduced species’ phenotypes fluctuate through time responding to changes in the selective forces, then a future direction would be determining what selection processes are the most common, as well as their relative importance across the process of invasion.

Two native species showed significant change through time, and in both cases the invasive species of the pair showed significantly higher rates of change than did the native species. That is, our data supports the idea that invasive species go through phenotypic change more often and faster than native species (Buswell et al., 2011).

171 Phenotypic change in invasive species

Both native and introduced species experience a wide range of selective forces, such as climate change, competition and land use change. However, because invasive species can have poor adaptation to the new environment (Broennimann et al., 2007; Gallagher et al., 2010), invasive species might be experiencing stronger selective pressures.

Another possibility is that invasive species are experiencing different selective forces than do native species. For instance, invasive species could be experiencing higher rates of predation or competition than the native species (Joshi and Vrieling, 2005; Parker et al., 2013).”

This study has provided important data on the trajectory of change in introduced species. The species we studied do not go through a lag phase in phenotypic change, and three out of three species show some evidence for change in at least one trait hundreds of years after their introduction. In particular, that introduced species can keep changing hundreds of years after being introduced suggests that some of them may yet reveal their full capacity in their novel ranges. However, the fluctuating nature of their phenotypic change suggests a lack of a constant strong selective force driving rapid directional change in introduced species. The patterns of change in introduced species might impact not only their spread and success, but may also impact their potential for diversification. Understanding the long term temporal patterns of phenotypic change in introduced species may help to understand their impacts and ecological mechanisms, as well as their evolutionary potential as a source for budding diversification.

Acknowledgments

172 Phenotypic change in invasive species

To the curators and herbaria collections from Natural History Museum, University of

Oxford, University of Cambridge, University of Sheffield, Liverpool Museum,

Manchester Museum and Royal Kew Botanical Gardens. H.F.M. wants to thank to Dr

Mark Spencer and Serena Marner for their patience and invaluable advice and to

Rhiannon Dalrymple for her thoughtful comments on previous versions of this manuscript. H.F.M. was supported by a scholarship from the Evolution & Ecology

Research Centre at UNSW and the Australian Research Council (DP 0984222). A.T.M. was supported by funding from the Australian Research Council (DP 0984222).

References

Aikio, S., Duncan, R. P. & Hulme, P. E. 2010. Lag-phases in alien plant invasions:

separating the facts from the artefacts. Oikos, 119: 370-378.

Biological Records Centre. 2013. Online Atlas of the British and Irish Flora.

Blossey, B. & Notzold, R. 1995. Evolution of increased competitive ability in invasive

nonindigenous plants: A hypothesis. Journal of Ecology, 83: 887-887.

Bone, E. & Farres, a. 2001. Trends and rates of microevolution in plants. Genetica, 112-

113: 165-82.

Bossdorf, O., Auge, H., Lafuma, L., Rogers, W. E., Siemann, E. & Prati, D. 2005.

Phenotypic and genetic differentiation between native and introduced plant

populations. Oecologia, 144: 1-11.

Broennimann, O., Treier, U. a., Müller-Schärer, H., Thuiller, W., Peterson, a. T. &

Guisan, a. 2007. Evidence of climatic niche shift during biological invasion.

Ecology Letters, 10: 701-9.

173 Phenotypic change in invasive species

Buswell, J. M., Moles, A. T. & Hartley, S. 2011. Is rapid evolution common in

introduced plant species? Journal of Ecology, 99: 214-224.

Carroll, S. P., Dingle, H. & Klassen, S. P. 1997. Genetic differentiation of fitness-

associated traits among rapidly evolving populations of the Soapberry bug.

Evolution, 51: 1182-1188.

Cazelles, B., Chavez, M., Berteaux, D., Ménard, F., Vik, J. O., Jenouvrier, S. &

Stenseth, N. C. 2008. Wavelet analysis of ecological time series. Oecologia,

156: 287-304.

Cornelissen, J. H. C., Lavorel, S., Garnier, E., Diaz, S., Buchmann, N., Gurvich, D. E.,

Reich, P. B., ter Steege, H., Morgan, H. D., van der Heijden, M. G. A., et al.

2003. A handbook of protocols for standardised and easy measurement of plant

functional traits worldwide. Australian Journal of Botany, 51: 335-380.

Crooks, J. A. 2005. Lag times and exotic species: The ecology. Ecoscience, 12: 316-

329.

Daehler, C. C. 2009. Short Lag Times for Invasive Tropical Plants : Evidence from

Experimental Plantings in Hawai’i. PLoS ONE, 4: e4462.

Daehler, C. C. & Strong, D. R. 1997. Reduced herbivore resistance in introduced

smooth cordgrass (Spartina alterniflora ) after a century of herbivore-free

growth. Oecologia, 110: 99-108.

Facon, B., Genton, B. J., Shykoff, J., Jarne, P., Estoup, A. & David, P. 2006. A general

eco-evolutionary framework for understanding bioinvasions. Trends in Ecology

& Evolution, 21: 130-5.

Felker-Quinn, E., Schweitzer, J. a. & Bailey, J. K. 2013. Meta-analysis reveals

evolution in invasive plant species but little support for Evolution of Increased

Competitive Ability (EICA). Ecology and Evolution, 3: 739-751.

174 Phenotypic change in invasive species

Gallagher, R. V., Beaumont, L. J., Hughes, L. & Leishman, M. R. 2010. Evidence for

climatic niche and biome shifts between native and novel ranges in plant species

introduced to Australia. Journal of Ecology, 98: 790-799.

Hendry, A. P. & Kinnison, M. T. 1999. The pace of modern life: measuring rates of

contemporary microevolution. Evolution, 53: 1637-1653.

Hendry, A. P., Nosil, P. & Rieseberg, L. H. 2007. The speed of ecological speciation.

Functional Ecology, 21: 455-464.

Hobbs, R. J. & Humphriest, S. E. 1995. An Integrated to the Ecology Approach

Invasions of Plant and Management. Conservation Biology, 9: 761-770.

Hoffmann, A. a. & Sgrò, C. M. 2011. Climate change and evolutionary adaptation.

Nature, 470: 479-85.

Holt, R. D., Barfield, M. & Gomulkiewicz, R. 2005. Theories of niche conservatism and

evolution: could exotic species be potential tests? In: D. F. Sax, J. J. Stachowicz

& S. D. Gaines (ed.), pp. pp. 259-290. Species invasions: insights inot ecology,

evolution, and biogeography, pp. pp. 259-290. Sinauer Associates, Sunderland,

MA.

Huey, R. B. 2000. Rapid evolution of a geographic cline in size in an introduced fly.

Science, 287: 308-309.

Joshi, J. & Vrieling, K. 2005. The enemy release and EICA hypothesis revisited:

incorporating the fundamental difference between specialist and generalist

herbivores. Ecology Letters, 8: 704-714.

Jump, A. S. & Penuelas, J. 2005. Running to stand still: adaptation and the response of

plants to rapid climate change. Ecology Letters, 8: 1010-1020.

Keane, R. M. & Crawley, M. J. 2002. Exotic plant invasions and the enemy release

hypothesis. Trends in Ecology & Evolution, 17: 164-170.

175 Phenotypic change in invasive species

Kowarik, I. 1995. Time lags in biological invasions. In: P. Pyšek, K. Prach, M.

Rejmánek & M. Wade (ed.), pp. 15-38. pp. 15-38. SPB Academic Publishing,

Amsterdam, Holland.

Lambrinos, J. G. 2004. How interactions between ecology and evolution influence

contemporary invasion dynamics. Ecology: 2061-2070.

Lee, C. E. 2002. Evolutionary genetics of invasive species. Trends in Ecology &

Evolution, 17: 9-11.

Mallat, S. 2008. A wavelet tour of signal processing: the sparse way. Academic Press,

San Diego, California, USA.

Mooney, H. A. & Cleland, E. E. 2001. The evolutionary impact of invasive species.

Sciences-New York, 98.

Orr, M. R. & Smith, T. B. 1998. Ecology and speciation. Trends in Ecology &

Evolution, 13: 502-506.

Parker, J. D., Torchin, M. E., Hufbauer, R. A., Lemoine, N. P., Alba, C., Blumenthal, D.

M., Bossdorf, O., Byers, J. E., Dunn, A. M., Heckman, R. W., et al. 2013. Do

invasive species perform better in their new ranges? Ecology, 94: 985-994.

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

of speed in toads. Nature, 439: 803-803.

Preston, C. D. & Hill, M. O. 1997. The geographical relationships of British and Irish

vascular plants. Botanical Journal of the Linnean Society, 124: 1-120.

Pyšek, P. & Prach, K. 1993. Plant invasions and the role of riparian habitats: a

comparison of four species alien to central Europe. Journal of Biogeography,

20: 413-420.

R Core Team. 2013. R: a language and environment for statistical computing. ed R

Foundation for Statistical Computing. Vienna, Austria.

176 Phenotypic change in invasive species

Rasband, W. S. 1997-2013. ImageJ, US National Institutes of Health, Bethesda,

Maryland, USA. National Institute of Health, Bethesda, Maryland.

Reznick, D. N. & Ghalambor, C. K. 2001. The population ecology of contemporary

adaptations: what empirical studies reveal about the conditions that promote

adaptive evolution. Genetica, 112-113: 183-98.

Sakai, A. K., Allendorf, F. W., Holt, J. S., Lodge, D. M., Molofsky, J., With, K. A.,

Baughman, S., Cabin, R. J., Cohen, J. E., Ellstrand, N. C., et al. 2001. The

population biology of invasive species. Annual Review of Ecology and

Systematics, 32: 305-332.

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

tropical Australia. Evolutionary Applications, 5: 107-116.

Strauss, S. Y. L. J. A. C. S. P. 2006. Evolutionary responses of natives to introduced

species : what do introductions tell us about natural communities ? Ecology

Letters, 9: 357-374.

Strong, D. R., Lawton, J. H. & Southwood, S. R. 1984. Insects on plants. Community

patterns and mechanisms. Harvard University Press, Cambridge, Massachusetts

USA.

Suarez, A. V. & Tsutsui, N. D. 2008. The evolutionary consequences of biological

invasions. Molecular Ecology, 17: 351-360.

Thompson, J. N. 1998. Rapid evolution as an ecological process. Trends in ecology &

evolution, 13: 329-332.

Vellend, M., Harmon, L. J., Lockwood, J. L., Mayfield, M. M., Hughes, a. R., Wares, J.

P. & Sax, D. F. 2007. Effects of exotic species on evolutionary diversification.

Trends in Ecology & Evolution, 22: 481-8.

177 Phenotypic change in invasive species

Westley, P. a. H. 2011. What invasive species reveal about the rate and form of

contemporary phenotypic change in nature. American Naturalist, 177: 496-509.

Westoby, M. 1998. A leaf-height-seed (LHS) plant ecology strategy scheme. Plant &

Soil, 199: 213-227.

Williams, C. K. & Moore, R. J. 1989. Phenotypic adaptation and natural selection in the

wild rabbit, Oryctolagus cuniculus, in Australia. Journal of Animal Ecology, 58:

495-507.

Williamson, M., Yo, Y. & Kingdom, U. 2005. On the rates and patterns of spread of

alien plants in the Czech Republic , Britain , and Ireland. Ecoscience, 12: 424-

433.

Zangerl, A. R., Stanley, M. C. & Berenbaum, M. R. 2008. Selection for chemical trait

remixing in an invasive weed after reassociation with a coevolved specialist.

Proceedings of the National Academy of Sciences, 105: 4547-4552.

Supporting information

Supporting information 1. Methodology to quantify the trajectory of the phenotypic change using wavelets.

Supporting information 2. Trajectory of the phenotypic change on the leaf area, leaf shape and plant height of three introduced species.

Supporting table 1. Comparison of the rate of phenotypic change through time (years) in three introduced species accounting for the effect of region.

178 Phenotypic change in invasive species

Supporting table 2. Comparison of the fit of linear and power models to the rate of phenotypic change through the first 100 years of introduction for three introduced species.

179 Phenotypic change in invasive species

Supporting information 1. Methodology to quantify the trajectory of the phenotypic change using wavelets.

The DWT analysis requires data sampled on regular time intervals, and herbarium specimens are not sampled on regular time intervals. Thus, we did a linear interpolation

trait trait trait for the time series (x = {x (t)t T }; where x denotes the time series related to species x for a given trait, t denotes the time stamp, and the set of available time stamps

T might be different for each species). We did not interpolate when a separation of >15 years occurred between points.

Then, we normalized each time series:

Equation 1 where is a normalized time series for a trait of species x, σtrait is the standard deviation for a given trait of species x, and is the mean value for a given trait of species x) to have zero mean and unit standard deviation. This normalization makes the trait data dimensionless allowing us to compare data between and within species.

Once we had the data normalized, we used DWT to decompose the data ( ) into

J levels, obtaining J sets of detailed coefficients (short term trends) at different resolutions (i.e. , where j denotes the decomposition level and is an appropriate set of indices) and one set of approximation coefficients (long

-J term trends) at a resolution 2 (i.e. , where the index J denotes the

180 Phenotypic change in invasive species

number decomposition levels). A normalized time series for a given species for a given trait ( trait), can be decomposed into scaling (Equation 2) and wavelet coefficients

(Equation 3):

Equation 2

Equation 3

In these equations the operator denotes the inner product in , and the scaling

function ( ) has the form and the wavelet function

( ) has the form . The decomposition level is related to the resolution while the index is related to the coarsest resolution

.

Within this framework, the reconstructed time series that only considers the low frequency components (or long term trends) at a resolution 2 J is calculated using

Equation 4, below:

Equation 4

For our time series we selected scaling and wavelet functions from the Daubechies family because they have the maximum number of vanishing moments for a given support, allowing the scaling and wavelet functions to characterize more complex functions(Mallat, 2008). We selected the wavelet db3 with J = 3 decomposition levels

181 Phenotypic change in invasive species

because of the number of coefficients of the functions and the reduced number of samples available for our experiment.

Once that we had the low frequencies (long trends) time series ( ) for each species we quantified changes in leaf area, leaf shape and plant height using the derivative from the filtered time series ( ). In the case of a discrete time series for a trait x the derivative is the difference between trait values in different years. We calculated the average phenotypic change as the average of the absolute values of this derivate using a sliding window (w; i.e. a time window frame of fixed size defined by two moving endpoints that move forward replacing old items as new data items arrive). The size of the sliding window w controls the time horizon on which the average phenotypic change is calculated. In our case w = 20 years, because although previous studies have reported a decade as a common minimum time period for phenotypic changes to occur (Bone and

Farres, 2001; Whitney and Gabler, 2008; Shine, 2012) timescales finer than this would have required too much interpolation. That is, we are evaluating the phenotypic change for each species across the time series respect to the past 20 years. The phenotypic change coefficient for the filtered time series can be expressed as:

Equation 5

We obtained a global phenotypic change coefficient for all the traits of a given species at a given time (t) by averaging the phenotypic change coefficients associated with each of our traits (i.e. leaf area, leaf shape and plant height):

Equation 6

182 Phenotypic change in invasive species

Since the phenotypic change coefficients and E(t) are obtained from normalized time series, we quantified changes in the rates of change by simply dividing the rates of phenotypic change observed in a focal time period (e.g. first decade after introduction) by any other time period of interest (e.g. second decade after introduction). All the analyses for describing the trajectories of phenotypic change using

DWT were done in Matlab v. 2010b.

References

Bone, E. & Farres, a. 2001. Trends and rates of microevolution in plants. Genetica, 112-

113: 165-82.

Mallat, S. 2008. A wavelet tour of signal processing: the sparse way. Academic Press,

San Diego, California, USA.

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

tropical Australia. Evolutionary Applications, 5: 107-116.

Whitney, K. D. & Gabler, C. A. 2008. Rapid evolution in introduced species , ‘invasive

traits’ and recipient communities: challenges for predicting invasive potential.

Diversity and Distributions: 569-580.

183 Phenotypic change in invasive species

Supporting information 2. Trajectory of the phenotypic change on the leaf area, leaf shape and plant height of three introduced species.

For leaf area Epilobium ciliatum shows rates of change from 0.001 to 0.21, with the maximum rate of change observed in 1927 (Figure 2). Senecio squalidus leaf area rates of change ranges from 0.07 to 0.29, with the maximum rate of change in 1907 (Figure

2). For Senecio squalidus leaf area was the trait with the highest rate of change.

Veronica persica’s leaf area rates of change ranges from 0.03 to 0.27, with the maximum rate of change in 1985 (Figure 2).

In the case of leaf shape Epilobium ciliatum shows rates of change between 0.01 and

0.21, with the maximum rate of change observed in 1934 (Figure 2). Senecio squalidus rate of change on leaf shape ranges from 0.02 to 0.22, with the maximum rate of change observed in 1907, as its’ leaf area does (Figure 2). For Veronica persica the rates of change on leaf shape ranged from 0.04 to 0.3, with the maximum rate of change observed in 2005 (Figure 2). For Veronica persica leaf shape was the trait that registered the highest rate of change.

For Epilobium ciliatum plant height rate of change varies from 0.005 to 0.31, with the maximum rate of change observed in 1930 (Figure 2). This is the trait that shows the highest rate of change for this species. In the case of Senecio squalidus the rate of change on plant height ranges from 0.04 to 0.24 (Figure 2). For this trait for this species the maximum rate of change was observed in 1980 (Figure 2). For Veronica persica the

184 Phenotypic change in invasive species

rates of change for plant height ranged from 0.07 to 0.16 (Figure 2). We observed the maximum rate of change on plant height for this species on 1937.

185 Phenotypic change in invasive species

Supporting table 1. Comparison of the rate of phenotypic change through time

(years) in three introduced species accounting for the effect of region.

1) Comparison of Epilobium ciliatum’s leaf area rate of change.

Terms Sum of squares d.f. F value P

Year 0.39 1 12.68 < 0.001

Region 0.07 1 2.12 0.15

Year × Region 0 1 0 0.99

Residuals 3.58 116

2) Comparison of Epilobium ciliatum’s leaf shape rate of change.

Terms Sum of squares d.f. F value P

Year 0.01 1 2.34 0.13

Region 0.0001 1 0.023 0.88

Year × Region 0.009 1 2.05 0.16

Residuals 0.53 116

186 Phenotypic change in invasive species

3) Comparison of Epilobium ciliatum’s plant height rate of change.

Terms Sum of squares d.f. F value P

Year 0.06 1 1.93 0.17

Region 0.002 1 0.09 0.77

Year × Region 0.004 1 0.15 0.7

Residuals 3.28 116

4) Comparison of Senecio squalidus’s leaf area rate of change.

Terms Sum of squares d.f. F value P

Year 0.32 1 5.42 0.02

Region 0.12 1 2.05 0.15

Year × Region 0.04 1 0.7 0.41

Residuals 11.65 197

5) Comparison of Senecio squalidus’s leaf shape rate of change.

Terms Sum of squares d.f. F value P

Year 0.005 1 0.37 0.55

Region 0.001 1 0.1 0.76

Year × Region 0.03 1 2.55 0.11

Residuals 2.62 197

187 Phenotypic change in invasive species

6) Comparison of Senecio squalidus’s plant height rate of change.

Terms Sum of squares d.f. F value P

Year 0.19 1 7.93 0.005

Region 0.0007 1 0.03 0.87

Year × Region 0.009 1 0.36 0.55

Residuals 4.76 197

7) Comparison of Veronica persica’s leaf area rate of change.

Terms Sum of squares d.f. F value P

Year 0.18 1 4.69 0.03

Region 0.003 1 0.07 0.8

Year × Region 0.02 1 0.57 0.45

Residuals 6.97 180

8) Comparison of Veronica persica’s leaf shape rate of change.

Terms Sum of squares d.f. F value P

Year 0.01 1 4.62 0.03

Region 0.002 1 0.98 0.32

Year × Region 0.0007 1 0.34 0.56

Residuals 0.37 180

188 Phenotypic change in invasive species

9) Comparison of Veronica persica’s plant height rate of change.

Terms Sum of squares d.f. F value P

Year 0.12 1 5.17 0.02

Region 0.004 1 0.15 0.70

Year × Region 0.004 1 0.16 0.7

Residuals 4.25 180

189 Phenotypic change in invasive species

Supporting table 2. Comparison of the fit of linear and power models to the rate of phenotypic change through the first 100 years of introduction for three introduced species.

1) Epilobium ciliatum leaf area

Model Residuals Residuals Sum Sum of d.f. of squares d.f. squares P Power Model 116 4.10 Linear Model 115 3.62 1 0.48 < 0.001

2) Epilobium ciliatum leaf shape

Model Residuals Residuals Sum of Sum of d.f. squares d.f. squares P Power Model 116 0.51 Linear Model 115 0.49 1 0.022 0.03

3) Epilobium ciliatum plant height

Model Residuals Residuals Sum Sum of d.f. of squares d.f. squares P Power Model 116 3.4 Linear Model 115 3.24 1 0.16 0.02

4) Senecio squalidus leaf area

Model Residuals Residuals Sum Sum of d.f. of squares d.f. squares P Power Model 46 3.66 Linear Model 45 3.31 1 0.34 0.04

190 Phenotypic change in invasive species

5) Senecio squalidus leaf shape

Model Residuals Residuals Sum Sum of d.f. of squares d.f. squares P Power Model 46 0.59 Linear Model 45 0.54 1 0.05 0.056

6) Senecio squalidus plant height

Model Residuals Residuals Sum Sum of d.f. of squares d.f. squares P Power Model 46 0.99 Linear Model 45 0.81 1 0.18 0.003

7) Veronica persica leaf area

Model Residuals Residuals Sum Sum of d.f. of squares d.f. squares P Power Model 125 4.92 Linear Model 124 4.62 1 0.30 0.006

8) Veronica persica leaf shape

Model Residuals Residuals Sum Sum of d.f. of squares d.f. squares P Power Model 125 0.22 Linear Model 124 0.21 1 0.009 0.03

9) Veronica persica plant height

Model Residuals Residuals Sum Sum of d.f. of squares d.f. squares P Power Model 125 2.87 Linear Model 124 2.71 1 0.15 0.009

191 Conclusion

In this thesis I asked whether introduced and native species differ in seed dispersal and recruitment success (chapters one and two) and found that on average introduced and native species do not significantly differ in these life history characteristics. Then I investigated the role of level of invasiveness (nature) and range shift (nurture) in species’ performance (chapter 3) and found that both nature and nurture influence invasive species’ seed production, but not dispersal or recruitment success. Finally, I investigated the evolutionary trajectories of invasive species shortly after their introduction (chapter four) and found that invasive species do not go through a lag phase and are capable of expressing significant phenotypic change hundreds of years after their introduction to a new range. In the following sections I discuss the importance of my findings.

Not all introduced species are super invaders

My results revealed a wide variation in seed dispersal and recruitment success in both introduced and native species (chapter one and two). That is, not all introduced species are good dispersers and not all of them have high recruitment success, and perhaps we shouldn’t expect them to be so. If introduced species have advantages over native species, such as enemy release (Keane and Crawley, 2002) or evolution of increased competitive ability (Blossey and Notzold, 1995), it is possible that this could be counterbalanced by native species’ superior adaptation to local conditions. For example, introduced species might have traits that are generally associated with longer seed

192 Conclusion

dispersal and higher recruitment success. However, seed dispersal may be heavily influenced by external factors like the availability of dispersers (Howe and Smallwood,

1982; Jongejans et al., 2008; Thomson et al., 2010), while recruitment success will vary depending on interactions with herbivores, pathogens and plant competitors and environmental factors such as temperature, water availability, or soil type (Kitajima and

Fenner, 2000; Moles and Westoby, 2006; Donohue et al., 2010). Overall, my results suggest that higher performance – or the perks of being introduced to a new range – do not generate seed dispersal or recruitment advantages in introduced species. We need to reassess our understanding of the ecological mechanisms behind introduced species’ success and accept the fact that native and introduced species are not as different as we first thought. My research has shown that it is unlikely that we will ever have the capability to reliably predict those species with a higher probability of success when introduced to in a new range solely based on traits like dispersal or recruitment. This strengthens the assertions that predictive models for invasive species may be a ‘holy grail’ for which we may long strive, but never satisfactorily achieve ( Williamson, 1999;

Davis et al., 2011). Promise, however, lies within one trait – seed production.

The chief role of seed production and the importance of range

Theory suggests that seed production is related to the success of introduced species

(Pyšek et al., 2012; Richardson and Pyšek, 2012). Mason et al. (2008) showed that on average invasive species have higher seed production than do native species. My results suggest that higher seed production is both a characteristic intrinsic to invasive species, and is positively affected by their new range (chapter 3). My findings are consistent with a recent meta-analysis by Parker et al. (2013) that shows that invasive species are

193 Conclusion

on average bigger and more fecund in their introduced range. The higher seed production of invasive species (Mason et al., 2008) could be directly associated with invasive species’ higher population growth rate (Ramula et al., 2008). The higher seed production of invasive species could be a result of benefits from diverse mechanisms, such as less herbivore and pathogen damage in reproductive parts (Liu and Stiling,

2006; but see Chun et al., 2010), evolutionary change in the allocation of resources from defence to reproduction (Keane and Crawley, 2002; Hierro et al., 2005; but see

Felker-Quinn et al., 2013), or greater biomass response to increase in resource availability (Dawson et al., 2012).

My work (chapter 3) was the first to quantify the relative importance of species’ level of invasiveness and species’ range on seed production across invasive species. My work shows that invasive species have higher seed production in their introduced range than they have in their native range. However, there are different ways that invasive species could achieve higher seed production. Species’ seed production is affected by access to pollinators, fruit and flower abortion and even the cost of the seed packaging

(Bierzychudek, 1981; Stephenson, 1981; Cavers, 1983). For example, a meta-analysis by Morales and Traveset (2009) that shows that on average introduced species have a negative effect on pollinator visitation and reproductive success of native species; also, a study by Rayamajhi et al. (2002) that shows that the invasive species Melaleuca quinquenervia has less aborted capsules in its’ introduced range. An important step forward in invasion ecology would be to investigate how introduced species achieve higher seed production. Quantifying how much of the demonstrated differences in seed production are related to more pollination vs. less accessory costs to reproduction is an important piece of understanding how introduced species are fitting into, and potentially

194 Conclusion

altering, the ecosystems of their new ranges. This thesis has demonstrated that simply observing one or a few species in one place at one time is not sufficient for deriving general rules– range and ecology, species interactions and environment dynamics are crucial for coming to understand fitness, function and success.

The tendency for introduced species to change their phenotypes over time (chapter 4) might relate to evolution of increased fitness in these landscapes, as they adapt to their new ranges. Having demonstrated the chief importance of seed production, we must draw these results together and come to the question: does invasive species’ seed production increase or decrease with residence time? Without advantages in recruitment success, the high population growth rates witnessed in many introduced species

(Ramula et al., 2008) may be due to their higher seed production alone. We must find out how these populations change in seed production and their demographic characteristics over time if we hope to model their continuing spread and prepare for the future environmental challenges that they may represent.

You can’t have it all at once: the costs of invasive species’ life history

Invasive species have higher seed production and bigger seed mass in their introduced range (chapter 3). However my study also shows that invasive species do not have higher recruitment in their introduced range (chapter 3). At first glance this seems to be contrary to the relationship between early seedling survival and seed mass (Moles and

Westoby, 2004). However, invasive species’ life history might have costs to fitness. For instance, invasive species seem to consistently show higher growth rates (Grotkopp and

Rejmánek, 2007; Dawson et al., 2011) and seed production than do non-invasive

195 Conclusion

species (Mason et al., 2008). Although these characteristics might help introduced species to achieve dominance and spread, higher growth rates might come with higher costs to survival (Metcalf et al., 2003; Turnbull et al., 2008; Rose et al., 2009).

Similarly, higher allocation of resources to reproductive output could have a negative effect on plant size (Bazzaz et al., 2000) of invasive species. Not only can the allocation of resources differ between species, but also timing for the allocation of resources to reproduction and vegetative growth may differ (Vincent and Brown, 1984; Kawecki,

1993; summarized in Falster and Westoby, 2003).

Important questions arise from these results. Quantifying whether the timing of allocation to reproduction and vegetative growth differs between invasive and non- invasive species is paramount. This is an important movement to understand the dynamics of invasion –away from a solely trait-centric approach. In addition, if we understand how introduced species allocate resources, their timing of their allocation, and the costs incurred in their ecological strategies, this could help us to understand the dynamics of successful colonizing species and the evolution of ecological strategies within plant communities.

The effects of publication bias.

All scientific disciplines, including ecology and evolution, are susceptible to publication bias (Møller and Jennions, 2001; Jennions and Møller, 2002). The existence of systematic bias in the literature could hamper the progress of ecology, particularly in synthesis work. For instance, publication bias can inflate the estimates of average effect sizes (Jennions and Møller, 2002), affecting general trends in results (Møller and

196 Conclusion

Jennions, 2001). If only studies with significant results are published this could result in the report of spurious trends in synthesis papers (Møller and Jennions, 2001). Levine et al. (2003) suggests that in invasion ecology, studies reporting significant invader impacts are more likely to be reported. Vila et al. (2011) reported a mild publication bias in favour of studies that show that invasive species have an ecological impact on the invaded communities. Similarly, van Kleunen et al. (2010) report that studies with small sample size are more likely to be reported when they show higher trait values for invasive species. Our study could have been affected by publication bias in the invasion literature, particular by studies comparing the traits of invasive and native species.

However, if this was the case the publication bias in favour of reporting invasive species outcompeting native ones (van Kleunen et al., 2010) would bias the comparison of life history characteristics between introduced/invasive and native species (chapter one, two and three) in favour of the introduced/invasive ones. That is, the publication bias in the invasive literature would only have increased the probability of finding significant differences between introduced and native species. Despite this possible bias, I found no difference in the dispersal distance (Chapter 1) and recruitment success of introduced and native species (Chapter 2), or even between invasive and non-invasive species

(Chapter 3).

The perils of proxies

My thesis highlights the importance of directly comparing process, such as seed dispersal or recruitment, rather than traits related to them (proxies), if information on processes is what is sought. Most of the time we assume that traits impact fitness, but we seldom know this for sure (Ackerly and Monson, 2003). The problem with the use

197 Conclusion

of traits as proxies for describing introduced species’ processes might have broader implications for the study of ecology as a whole.

Currently, many areas in ecology, such as community ecology, invasion ecology, behavioural ecology and phylogenetic ecology, are using trait data for inferring ecological processes. This is related to the fact that we now have trait databases containing millions of records from all over the world (see Kattge et al., 2011).

However, as this thesis shows, many problems arise with the use of traits; it can lead to ecological conundrums, arising from the inference of information from trait data, rather than from the deceiving nature of ecological mechanism and processes. For instance, the confounding results emerging from the comparison of some traits between introduced and native species might be reflecting that these traits are uninformative to the understanding of competitive interactions in communities.

Currently the overwhelming use of traits limits our power to understand ecological processes and mechanisms. In this respect, measuring the relationship between demographic rates and traits stands as an important line of research (Poorter and

Bongers, 2006; Martínez-Vilalta et al., 2010). Another essential area of research is the systematic collection and compilation of demographic information at a global scale.

This mammoth task has begun; for instance, the ComPADRe III database (Comparative

Plant and Algae Demographic Research, Salguero-Gomez, unpublished data) contains demographic information on hundreds of plants around the world.

198 Conclusion

Species can change, and they keep doing it

While we know that introduced species can change through time (Thompson, 1998;

Bone and Farres, 2001; Lambrinos, 2004; Buswell et al., 2011), we know little about the pattern of this change. I found that introduced species do not seem to go through a lag phase and that hundreds of years after introduction invasive species are still capable of rapid phenotypic change (chapter four).

Rapid periods of strong selection can influence invasive species, but the strength of selective forces will change through time. Early in an invasion, introduced species might be less exposed to herbivores, allowing allocation of defence resources to growth or reproduction (Blossey and Notzold, 1995; Keane and Crawley, 2002), however, as the population increases in range and abundance, it will be more exposed to herbivores

(Strong et al., 1984). An interesting, consequent future research direction would be to figure out the relative importance of different selective processes through the colonization, establishment and spread of invasive species. That introduced species change continuously over long time periods indicates that, given adequate time, reproductively isolated populations could change in their novel ranges sufficiently to ecologically diverge and speciate (Vellend et al., 2007; Moles et al., 2012).

Investigating how fast and how often invasive species are going through the process of ecological speciation is an exciting next move – crucial to understanding the long term patterns that invasive species may undergo. Controversially, if these species do in fact represent examples of contemporary speciation in process, then the value of extirpation measures could be called into question.

199 Conclusion

Practical implications

Overall this thesis provides good news to managers. First of all, not all invasive species are super invaders (chapter one and two). More time and resources should be allocated to species that do impact ecosystems, rather than all invasive species (Davis et al.,

2011), as not only is it logistically impossible to attack all invasive species, but costly in both time and money. Let’s work smart, not hard. Second, invasive species are not always better throughout the different stages of the life cycle (chapter three). This is important as it suggests that even among successful introduced species there are crucial stages where more management efforts should be focused in order to obtain better results. For instance, management plans in Australia have shifted their focus from introduced species in general to managing the most invasive ones or the ones in the risk to become widespread (Thorp et al., 2000). However, still much information is needed as many claims on the threats that introduced species represent are groundless

(Gurevitch and Padilla, 2004; Davis et al., 2011). My study showed that seed production is the only phase during which introduced species generally have an advantage over native species. The importance of seed production for the control of invasive species has also been observed in species-specific studies (Davis et al., 2006;

Jongejans et al., 2006). Third, introduced species do not go through a lag phase and are still changing hundreds of years after their introduction. It is possible that we are yet to see the full potential of some introduced species as invaders. However, this result is mainly encouraging as it suggests that species are more evolutionarily labile than we thought and may be better prepared to face sudden environmental changes.

Synthesis

200 Conclusion

Many studies have found that invasive and non-invasive species differ in their traits

(Pyšek and Richardson, 2007; van Kleunen et al., 2010; Richardson and Pyšek, 2012).

To list just three examples, many studies have found differences in plant height

(Ordonez et al., 2010), seed mass (Buckley et al., 2003; Ordonez et al., 2010) and leaf mass per area (Leishman et al., 2007) between invasive and non-invasive species.

However trait differences between species tend to be a weak predictor of invasiveness.

For instance in chapters one, two and three of this thesis I showed that invasive and non-invasive species tend to differ in important ecological traits like seed mass and plant height (Westoby et al., 2002; Moles and Westoby, 2004; Moles and Westoby,

2006; Moles et al., 2009). However they do not differ in recruitment success or seed dispersal distance, as previously was suggested (Rejmanek and Richardson, 1996;

Colautti et al., 2006; Pyšek and Richardson, 2007; Murray and Phillips, 2010; Ordonez and Olff, 2013). Life history characteristics vary across invasive species as much as they do in native species. This could explain why invasiveness cannot be consistently predicted from common life history characteristics. Nevertheless, seed production is particularly informative for understanding species’ levels of invasiveness (chapter three). To increase our predictive power in determining species’ level of invasiveness, we need to consider multiple variables, both intrinsic and extrinsic to the species.

Considering variables that directly affect survival, growth and reproduction may increase our power to predict invasiveness.

Summary

It has been long proposed that introduced/invasive species outcompete and outperform native or non-invasive species (Lodge, 1993; Sakai et al., 2001; van Kleunen et al.,

201 Conclusion

2010).This thesis shows that, in fact, introduced and native species are not that different in important life history characteristics at a broad scale under natural conditions.

Furthermore, these results show that that the use of traits as proxies in invasive species can be misguiding and sometimes misinforming for describing ecological processes.

This finding impacts not only research in invasion ecology, but ecology as a whole.

Invasion ecology has long pushed the bounds - integrating knowledge and evidence from many walks of ecological and evolutionary theory. My contribution is a meaningful step on the path, along which invasion ecology continues to advance our methodology and general ecological theory.

References

Ackerly, D. D. & Monson, R. K. 2003. Waking the Sleeping Giant : the evolutionary

foundations of plant function. International Journal of Plant Sciences, 164: S1-

S6.

Bazzaz, F. A., Ackerly, D. D. & Reekie, E. G. 2000. Reproductive allocation in plants.

In: M. Fenner (ed.), pp. pp. 1-29. pp. pp. 1-29. CAB International, Wallingford.

Bierzychudek, P. 1981. Pollinator limitation of plant reproductive effort. American

Naturalist, 117: 838-840.

Blossey, B. & Notzold, R. 1995. Evolution of increased competitive ability in invasive

nonindigenous plants: A hypothesis. Journal of Ecology, 83: 887-887.

Bone, E. & Farres, a. 2001. Trends and rates of microevolution in plants. Genetica, 112-

113: 165-82.

Buckley, Y. M., Downey, P., Fowler, S. V., Hill, R., Memmot, J., Norambuena, H.,

Pitcairn, M., Shaw, R., Sheppard, A. W., Winks, C., et al. 2003. Are invasive

202 Conclusion

bigger? A global study of seed size variation in two invasive shrubs. Ecology,

84: 1434-1440.

Buswell, J. M., Moles, A. T. & Hartley, S. 2011. Is rapid evolution common in

introduced plant species? Journal of Ecology, 99: 214-224.

Cavers, P. B. 1983. Seed demography. Canadian Journal of Botany, 61: 3578-3590.

Chun, Y. J., van Kleunen, M. & Dawson, W. 2010. The role of enemy release, tolerance

and resistance in plant invasions: linking damage to performance. Ecology

Letters, 13: 937-946.

Colautti, R. I., Grigorovich, I. A. & MacIsaac, H. J. 2006. Propagule pressure: a null

model for biological invasions. Biological Invasions, 8: 1023-1037.

Davis, A. S., Landis, D. a., Nuzzo, V., Blossey, B., Gerber, E. & Hinz, H. L. 2006.

Demographic models inform selection of biocontrol agents for garlic mustard

(Alliaria petiolata). Ecological Applications, 16: 2399-410.

Davis, M. A., Chew, M. K., Hobbs, R. J., Lugo, A. E., Ewel, J. J., Vermeij, G. J.,

Brown, J. H., Rosenzweig, M. L., Gardener, M. R., Carroll, S. P., et al. 2011.

Don't judge species on their origins. Nature, 474: 153-4.

Dawson, W., Fischer, M. & van Kleunen, M. 2011. The maximum relative growth rate

of common UK plant species is positively associated with their global

invasiveness. Global Ecology and Biogeography, 20: 299-306.

Dawson, W., Rohr, R. P., van Kleunen, M. & Fischer, M. 2012. Alien plant species with

a wider global distribution are better able to capitalize on increased resource

availability. New Phytologist, 194: 859-67.

Donohue, K., Rubio de Casas, R., Burghardt, L., Kovach, K. & Willis, C. G. 2010.

Germination, Postgermination Adaptation, and Species Ecological Ranges.

Annual Review of Ecology, Evolution, and Systematics, 41: 293-319.

203 Conclusion

Falster, D. S. & Westoby, M. 2003. Plant height and evolutionary games. Trends in

Ecology & Evolution, 18: 337-343.

Felker-Quinn, E., Schweitzer, J. a. & Bailey, J. K. 2013. Meta-analysis reveals

evolution in invasive plant species but little support for Evolution of Increased

Competitive Ability (EICA). Ecology and Evolution, 3: 739-751.

Grotkopp, E. & Rejmánek, M. 2007. High seedling relative growth rate and specific leaf

area are traits of invasive species: Philogenetically independent contrasts of

woody angiosperms. American Journal of Botany, 94: 526-532.

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

Trends in ecology & evolution (Personal edition), 19: 470-4.

Hierro, J. L., Maron, J. L. & Callaway, R. M. 2005. A biogeographical approach to

plant invasions : the importance of studying exotics in their introduced and

native range. Journal of Ecology: 5-15.

Howe, H. F. & Smallwood, J. 1982. Ecology of seed dispersal. Annual Review of

Ecology and Systematics, 13: 201-228.

Jennions, M. D. & Møller, A. P. 2002. Publication bias in ecology and evolution: an

empirical assessment using the 'trim and fill' method. Biological reviews of the

Cambridge Philosophical Society, 77: 211-22.

Jongejans, E., Shea, K., Skarpaas, O., Kelly, D., Sheppard, A. W. & Woodburn, T. L.

2008. Dispersal and demography contributions to population spread of Carduus

nutans in its native and invaded ranges. Journal of Ecology, 96: 687-697.

Jongejans, E., Sheppard, A. W. & Shea, K. 2006. What controls the population

dynamics of the invasive thistle Carduus nutans in its native range? Journal of

Applied Ecology, 43: 877-886.

204 Conclusion

Kattge, J., Díaz, S., Lavorel, S., Prentice, I. C., Leadley, P., Bönisch, G., Garnier, E.,

Westoby, M., Reich, P. B., Wright, I. J., et al. 2011. TRY - a global database of

plant traits. Global Change Biology, 17: 2905-2935.

Kawecki, T. J. 1993. Age and size at maturity in a patchy environment: fitness

maximization versus evolutionary stability. Oikos.

Keane, R. M. & Crawley, M. J. 2002. Exotic plant invasions and the enemy release

hypothesis. Trends in Ecology & Evolution, 17: 164-170.

Kitajima, K. & Fenner, M. 2000. Ecology of seedling regeneration. In: M. Fenner (ed.),

pp. 331-359. Seeds: The ecology of regeneration in plant communities, pp. 331-

359. CABI Publishing, New York.

Lambrinos, J. G. 2004. How interactions between ecology and evolution influence

contemporary invasion dynamics. Ecology: 2061-2070.

Leishman, M. R., Haslehurst, T., Ares, A. & Baruch, Z. 2007. Leaf trait relationships of

native and invasive plants: community- and global-scale comparisons. New

Phytologist, 176: 635-43.

Levine, J. M., Vilà, M., D'Antonio, C. M., Dukes, J. S., Grigulis, K. & Lavorel, S. 2003.

Mechanisms underlying the impacts of exotic plant invasions. Proceedings.

Biological sciences / The Royal Society, 270: 775-81.

Liu, H. & Stiling, P. 2006. Testing the enemy release hypothesis: a review and meta-

analysis. Biological Invasions, 8: 1535-1545.

Lodge, D. M. 1993. Biological Invasions: lessons for Ecology. Trends in Ecology &

Evolution, 8: 133-137.

Martínez-Vilalta, J., Mencuccini, M., Vayreda, J. & Retana, J. 2010. Interspecific

variation in functional traits, not climatic differences among species ranges,

205 Conclusion

determines demographic rates across 44 temperate and Mediterranean tree

species. Journal of Ecology, 98: 1462-1475.

Mason, R. A. B., Cooke, J., Moles, A. T. & Leishman, M. R. 2008. Reproductive output

of invasive versus native plants. Global Ecology and Biogeography, 17: 633-

640.

Metcalf, J. C., Rose, K. E. & Rees, M. 2003. Evolutionary demography of monocarpic

perennials. Trends in Ecology & Evolution, 18: 471-480.

Moles, A. T., Flores-Moreno, H., Bonser, S. P., Warton, D. I., Helm, A., Warman, L.,

Eldridge, D. J., Jurado, E., Hemmings, F. A., Reich, P. B., et al. 2012. Invasions:

the trail behind, the path ahead, and a test of a disturbing idea. Journal of

Ecology, 100: 116-127.

Moles, A. T., Warton, D. I., Warman, L., Swenson, N. G., Laffan, S. W., Zanne, A. E.,

Pitman, A., Hemmings, F. a. & Leishman, M. R. 2009. Global patterns in plant

height. Journal of Ecology, 97: 923-932.

Moles, A. T. & Westoby, M. 2004. Seedling survival and seed size: a synthesis of the

literature. Journal of Ecology, 92: 372-383.

Moles, A. T. & Westoby, M. 2006. Seed size and plant strategy across the whole life

cycle. Oikos, 113: 91-105.

Møller, A. P. & Jennions, M. D. 2001. Testing and adjusting for publication bias.

Trends in Ecology & Evolution, 16: 580-586.

Morales, C. L. & Traveset, A. 2009. A meta-analysis of impacts of alien vs. native

plants on pollinator visitation and reproductive success of co-flowering native

plants. Ecology Letters, 12: 716-728.

206 Conclusion

Murray, B. R. & Phillips, M. L. 2010. Investment in seed dispersal structures is linked

to invasiveness in exotic plant species of south-eastern Australia. Biological

Invasions, 12: 2265-2275.

Ordonez, A. & Olff, H. 2013. Do alien plant species profit more from high resource

supply than natives? A trait-based analysis. Global Ecology and Biogeography,

22: 648-658.

Ordonez, A., Wright, I. J. & Olff, H. 2010. Functional differences between native and

alien species: a global-scale comparison. Functional Ecology, 24: 1353-1361.

Parker, J. D., Torchin, M. E., Hufbauer, R. A., Lemoine, N. P., Alba, C., Blumenthal, D.

M., Bossdorf, O., Byers, J. E., Dunn, A. M., Heckman, R. W., et al. 2013. Do

invasive species perform better in their new ranges? Ecology, 94: 985-994.

Poorter, L. & Bongers, F. 2006. Leaf Traits Are Good Predictors of Plant Performance

across 53 Rain Forest Species. Ecology, 87: 1733-1743.

Pyšek, P., Daniehlka, J., Sadlo, J., Chrtek Jr, J., Chytry, M., Jarošík, V., Kaplan, Z.,

Krahulec, F., Moravcová, L., Pergl, J., et al. 2012. Catalogue of alien plants of

the Czech Republic (2nd edition): checklist update , taxonomic diversity and

invasion patterns. Preslia, 84: 155-255.

Pyšek, P. & Richardson, D. M. 2007. Traits associated with invasiveness in alien plants:

where do we stand? In: W. Nentwig (ed.), pp. 97-125. Biological Invasions,

Ecological Studies Vol. 193, pp. 97-125. Springer-Verlag, Berlin.

Ramula, S., Knight, T. M., Burns, J. H. & Buckley, Y. M. 2008. General guidelines for

invasive plant management based on comparative demography of invasive and

native plant populations. Journal of Applied Ecology, 45: 1124-1133.

207 Conclusion

Rayamajhi, M. B., Van, T. K., Goolsby, J. A., Pratt, P. D. & Racelis, A. 2002.

Biological Attributes of The Canopy-held Melaleuca Seeds in Australia and

Florida, U.S. Journal of Aquatic Plant Management, 40: 87-91.

Rejmanek, M. & Richardson, D. M. 1996. What attributes make some plant species

more invasive? Ecology, 78: 171-181.

Richardson, D. M. & Pyšek, P. 2012. Naturalization of introduced plants: ecological

drivers of biogeographical patterns. New Phytologist, 196: 383-96.

Rose, K. E., Atkinson, R. L., Turnbull, L. a. & Rees, M. 2009. The costs and benefits of

fast living. Ecology Letters, 12: 1379-84.

Sakai, A. K., Allendorf, F. W., Holt, J. S., Lodge, D. M., Molofsky, J., With, K. A.,

Baughman, S., Cabin, R. J., Cohen, J. E., Ellstrand, N. C., et al. 2001. The

population biology of invasive species. Annual Review of Ecology and

Systematics, 32: 305-332.

Stephenson, A. G. 1981. Flower and fruit abortion: proximate causes and ultimate

functions. Annual Review of Ecology and Systematics, 12: 253-279.

Strong, D. R., Lawton, J. H. & Southwood, S. R. 1984. Insects on plants. Community

patterns and mechanisms. Harvard University Press, Cambridge, Massachusetts

USA.

Thompson, J. N. 1998. Rapid evolution as an ecological process. Trends in ecology &

evolution, 13: 329-332.

Thomson, F. J., Moles, A. T., Auld, T. D., Ramp, D., Ren, S. & Kingsford, R. T. 2010.

Chasing the unknown: predicting seed dispersal mechanisms from plant traits.

Journal of Ecology, 98: 1310-1318.

Thorp, J. R., Lynch, R. & Trust, N. H. 2000. The determination of weeds of national

significance.

208 Conclusion

Turnbull, L. a., Paul-Victor, C., Schmid, B. & Purves, D. W. 2008. Growth rates, seed

size, and physiology: do small-seeded species really grow faster? Ecology, 89:

1352-63. van Kleunen, M., Weber, E. & Fischer, M. 2010. A meta-analysis of trait differences

between invasive and non-invasive plant species. Ecology Letters, 13: 235-245.

Vellend, M., Harmon, L. J., Lockwood, J. L., Mayfield, M. M., Hughes, a. R., Wares, J.

P. & Sax, D. F. 2007. Effects of exotic species on evolutionary diversification.

Trends in Ecology & Evolution, 22: 481-8.

Vilà, M., Espinar, J. L., Hejda, M., Hulme, P. E., Jarošík, V., Maron, J. L., Pergl, J.,

Schaffner, U., Sun, Y. & Pyšek, P. 2011. Ecological impacts of invasive alien

plants: a meta-analysis of their effects on species, communities and ecosystems.

Ecology Letters: 1-7.

Vincent, T. L. & Brown, J. S. 1984. Stability in an evolutionary game. Theoretical

Population Biology.

Westoby, M., Falster, D. S., Moles, A. T., Vesk, P. A. & Wright, I. J. 2002. Plant

ecological strategies: some leading dimensions of variation between species.

Annual Review of Ecology and Systematics, 33: 125-159.

Williamson, M. H. 1999. Invasions. Ecography, 5: 5-12.

209