The effects of herbivory and competition on 'Senecio inaequidens' DC. (Asteraceae), an invasive alien plant

Contents

Abstract

Zusammenfassung

Acknowledgements

Declaration

Preface

List of Figures

List of Plates

1 Introduction
1.1 Preliminary remarks
1.2 Why study biological invasions?
1.2.1 The borderless world
1.2.2 Biological invasions: concepts and definitions
1.2.3 Invasive plants
1.2.3.1 General overview
1.2.3.2 Important invasive alien plants in Europe
1.3 Herbivory and plant competition
1.3.1 Herbivory
1.3.2 Plant competition
1.4 The special case of Senecio inaequidens DC
1.4.1 The main reasons for focussing on this species
1.4.2 Phylogenetic position
1.4.3 Country of origin and natural habitats
1.4.4 Invasion history
1.4.4.1 World (excluding Eurasia)
1.4.4.2 Eurasia
1.4.4.3 Germany
1.4.4.4 United Kingdom
1.4.5 Biological characteristics
1.4.5.1 Morphology and life form
1.4.5.2 Reproductive biology
1.4.5.3 Herbivore spectrum and pathogens
1.4.5.4 Physiology, biochemistry and secondary metabolites
1.4.5.5 Plant sociology
1.4.6 Possible ecological impact
1.4.7 Possible economical impact
1.5 Current research and the objectives of this study
1.5.1 Overview
1.5.2 Hypotheses

2 Material and Methods
2.1 Living organisms and how they were obtained
2.1.1 Plants
2.1.1.1 Senecio inaequidens DC. (Asteraceae)
2.1.1.2 Festuca rubra L. ssp. rubra (Poaceae)
2.1.2 Herbivores
2.1.2.1 Tyria jacobaeae L. (Lepidoptera: Arctiidae)
2.1.2.1.1 Biology of the species
2.1.2.1.2 Experiments
2.1.2.2 Longitarsus jacobaeae Waterhouse (Coleoptera: Chrysomelidae)
2.1.2.2.1 Biology
2.1.2.2.2 Experiments
2.1.2.3 Oryctolagus cuniculus L. (Mammalia: Lagomorpha) and molluscs
2.2 Experimental design
2.2.1 Overview
2.2.2 Germination tests
2.2.3 Feeding trials
2.2.3.1 Feeding trials with Longitarsus jacobaeae
2.2.3.1.1 No-choice feeding preference experiment
2.2.3.1.2 Multiple-choice feeding preference experiments
2.2.3.2 Feeding trials with Tyria jacobaeae
2.2.3.2.1 No-choice feeding preference experiments
2.2.3.2.2 Multiple choice feeding preference experiments
2.2.4 Greenhouse experiments
2.2.5 Field experiments
2.2.5.1 Overview
2.2.5.2 Description of the field plots
2.2.5.3 Details on the split-plot design
2.2.5.4 Notes on the response variables used
2.2.6 Additional observations
2.2.6.1 Studies on natural populations
2.2.6.2 Pollination experiments
2.2.6.3 Vegetative reproduction
2.2.6.4 Test for vesicular-arbuscular mycorrhiza
2.3 Statistical Analysis

3 Results
3.1 Germination tests
3.2 Feeding trials
3.2.1 Longitarsus jacobaeae
3.2.1.1 No-choice feeding preference experiment
3.2.1.2 Multiple choice feeding preference experiments
3.2.1.2.1 Statistical analysis
3.2.1.2.2 Results
3.2.2 Tyria jacobaeae
3.2.2.1 Oviposition
3.2.2.2 Development and survival of larvae in no-choice tests
3.2.2.3 Damage to Senecio inaequidens in no-choice tests
3.2.2.4 Multiple choice feeding preference experiments
3.3 Greenhouse experiments
3.3.1 Plant competition
3.3.1.1 Statistical Analysis
3.3.1.2 Results
3.3.2 Herbivory
3.4 Field experiments
3.4.1 Statistical analysis
3.4.1.1 General description
3.4.1.2 Split-plot ANOVA involving rabbit grazing, plant competition, and ecotype effects
3.4.2 Frequency distributions of plant height in Senecio inaequidens
3.4.3 Differences between the two experimental plots
3.4.4 Differences between ecotypes of Senecio inaequidens
3.4.4.1 Ecotype differences in morphological parameters
3.4.4.2 Ecotype differences in invertebrate herbivore damage
3.4.4.3 Treatment effects
3.4.5 Direct effects of plant competition
3.4.6 Direct effects of herbivory
3.4.6.1 Vertebrate herbivores
3.4.6.2 Effects of invertebrate herbivores
3.4.7 Interaction effects
3.4.7.1 Interactions involving invertebrate herbivores
3.4.7.2 Interactions involving treatment and ecotype
3.4.7.3 Interactions between rabbit grazing and interspecific plant competition
3.4.8 Regrowth after clipping
3.4.9 Colonization by Longitarsus jacobaeae
3.4.10 Colonization by other insect herbivores
3.5 Additional observations
3.5.1 Studies on natural populations
3.5.2 Pollination experiments
3.5.3 Vegetative reproduction
3.5.4 Test for vesicular-arbuscular mycorrhiza

4 Discussion
4.1 Introductory remarks
4.2 Feeding trials
4.2.1 Field and laboratory trials with Longitarsus jacobaeae
4.2.2 Feeding trials with Tyria jacobaeae
4.3 Greenhouse experiments
4.4 Field experiments
4.4.1 Differences in morphological parameters, and between experimental plots
4.4.2 Differences between ecotypes of Senecio inaequidens
4.4.3 Herbivory and competition effects
4.5 Host-switching in oligophagous insect herbivores
4.6 Susceptibility of grassland ecosystems towards invasion by Senecio inaequidens

5 Final conclusions and need for further research

6 References

Abstract

Since the end of the 19th century, overall per-capita mobility of humans has increased significantly, leading to increased rates in human-mediated transportation of animal and plant species. The rapid spread of alien organisms, however, may lead to quick and unpredictable changes in ecosystems.

Senecio inaequidens DC. (Asteraceae) is an invasive alien plant from South Africa that was first introduced to Europe 100 years ago and is characterized by an exceptionally fast rate of spread; it contains pyrrolizidine alkaloids that are toxic to invertebrates, livestock and humans.

In the study presented here, field, laboratory and greenhouse experiments on the biology of Senecio inaequidens were conducted, in order to find out if and how herbivory and plant competition influence growth, survival and reproduction of this plant. Fully factorial or split-plot designs were used in order to answer this question. Vertebrate herbivores (rabbits) as well as native invertebrate herbivores (molluscs and two oligophagous insect herbivores) were used in the field experiments; the insect herbivores studied in detail were Longitarsus jacobaeae Waterhouse (Coleoptera: Chrysomelidae) and Tyria jacobaeae (Lepidoptera: Arctiidae). Multiple- and no-choice feeding preference experiments with these two insect herbivores were conducted. All experiments were performed at Imperial College, Silwood Park, about 30 km west of London (UK).

Growth and fecundity of Senecio inaequidens were significantly affected by interspecific competition, both in the greenhouse and in the field experiments. Rabbits only showed a significant effect when a closed vegetation cover was present at the same time. On artificially disturbed plots, Senecio inaequidens showed high capability to overcompensate for clipping of the above-ground parts by these herbivores. The shoots produced after clipping were not eaten by the rabbits any more, and the plants reached the flowering and fruit production stage within the same growth period.

There was a significant negative relationship between mollusc damage to leaves, and number of capitulae produced. Different ecotypes of Senecio inaequidens showed different amounts of herbivore damage.

One of the most remarkable results of this study was that Longitarsus jacobaeae freely colonized Senecio inaequidens, and that this plant was also accepted as a food source in laboratory two-way feeding preference experiments. In contrast, it could be clearly shown that Tyria jacobaeae does not accept Senecio inaequidens as a host plant.

Host switching from indigenous to invasive alien plant species in oligophagous insect herbivores may be more common than generally thought. The study presented here gives first experimental support for this assertion.

From the results of the experiments on rabbit grazing and plant competition, it is likely that Senecio inaequidens will start to colonize heavily grazed or disturbed grassland ecosystems in the near future; because of the toxic compounds this plant contains, it would be advisable to design a preventive management programme and to inform the public about the consequences that might be associated with the invasion of this plant.

Zusammenfassung

Die während des letzten Jahrhunders stark gestiegene Mobilität von Teilen der Weltbevölkerung hat dazu geführt, daß sich Tier- und Pflanzenarten über weitaus größere Distanzen ausbreiten können, als ihnen dies unter normalen Umständen möglich wäre. Die Verschleppung und Einbürgerung von Tier- und Pflanzenarten kann zu raschen und unvorhersehbaren Veränderungen in ökosystemen führen.

Senecio inaequidens DC. (Asteraceae) ist ein vor ca. hundert Jahren aus Südafrika eingeschleppter Neophyt, der durch eine besonders hohe Ausbreitungsgeschwindigkeit charakterisiert ist; die Pflanze enthält Pyrrolizidin-Alkaloide, die giftig für Wirbellose, Weidevieh und Menschen sind.

Im Rahmen der vorliegenden Arbeit wurden Experimente zur Biologie dieses Neophyten durchgeführt, mit dem Ziel, Prognosen über den zukünftigen Verlauf der Invasion in Abhängigkeit verschiedener Faktoren machen zu können. Hierzu wurden Gewächshaus-, Labor- und Freilandexperimente durchgeführt.

Die Auswirkungen von Herbivorie und interspezifischer Konkurrenz auf Wachstum und Reproduktion von Senecio inaequidens wurden in faktoriellen Experimenten und in sogenannten ‘split plot’ Experimenten untersucht. Als Herbivoren dienten Kaninchen, Schnecken und zwei auf eine andere Senecio -Art spezialisierte phytophage Insekten (Longitarsus jacobaeae Waterhouse, Coleoptera: Chrysomelidae und Tyria jacobaeae L., Lepidoptera: Arctiidae). Um zu testen, ob einheimische Herbivoren den Neophyten Senecio inaequidens als sekundäre Futterpflanze annehmen, wurden Fütterungsexperimente unter kontrollierten Laborbedingungen durchgeführt. Alle Untersuchungen fanden am Imperial College (Großbritannien) auf einem Forschungsgelände 30 km westlich von London statt.

Wachstum und Reproduktion von Senecio inaequidens wurden sowohl im Gewächshaus, als auch im Freiland signifikant durch interspezifische Konkurrenz beeinflußt. Kaninchen hatten nur dann einen signifikanten Effekt, wenn gleichzeitig eine geschlossene Vegetationsdecke vorhanden war. Bei gestörter Vegetationsdecke trat Regenerationswachstum an von Kaninchen dekapitierten Senecio inaequidens – Pflanzen auf, und die neu ausgetriebenen oberirdischen Organe wurden nicht mehr von den Kaninchen gefressen. Pflanzen, die auf diese Weise neu ausgetrieben waren, kamen noch im selben Jahr zur Blüte und Fruchtreife.

Je mehr Schneckenfraß an den Blättern auftrat, desto geringer war die Anzahl der produzierten Blütenköpfe. Verschiedene ökotypen von Senecio inaequidens wurden unterschiedlich stark von Longitarsus und Molluscen angenommen. Besonders bemerkenswert war, daß Longitarsus jacobaeae das Schmalblättrige Greiskraut von selbst zu besiedeln begann und dieses auch in Fraßtests unter Laborbedingungen als Futterquelle akzeptierte. Dagegen konnte klar gezeigt werden, daß Tyria jacobaeae diese Greiskraut-Art nicht als Futterpflanze annimmt.

Futterpflanzen-Wechsel von einheimischen zu eingeschleppten Pflanzenarten könnte bei spezialisierten phytophagen Insekten häufiger auftreten als bisher angenommen. Die vorliegende Arbeit liefert hierfür am Beispiel von Senecio inaequidens einen ersten Beleg.

Wie aus den Ergebnissen dieser Arbeit geschlussfolgert werden kann, ist es wahrscheinlich, daß Senecio inaequidens in naher Zukunft verstärkt gestörte und stark beweidete Graslandökosysteme besiedeln wird. Aufgrund der toxischen Inhaltsstoffe dieser Pflanze ist daher möglicherweise ein präventives Managementprogramm und die Aufklärung der Bevölkerung ratsam.

Acknowledgements

This work would not have been possible without the encouragement and support of my two supervisors, Professor Mick Crawley (Imperial College, Silwood Park, United Kingdom) and Professor Stefan Porembski (University of Rostock, Germany).

I would furthermore wish to thank all the people who have supported me in many ways: My parents, Jörg and Gertraude Scherber (München, Germany); Julia Wolf (Rostock and Jena, Germany) for all her love and support; the technicians and gardeners at Silwood Park, especially Jim Culverhouse, Pete Wilkinson and Paul Beasley; Dr. Johannes D. Nauenburg (Rostock Botanical Gardens) and Johannes Betz (Augsburg) for organizing seed deliveries for me; Dr. Christel Baum (Agrarwissenschaftliche Fakultät, Universität Rostock, Germany) for conducting the screening for VA mycorrhiza; Dr. Gregor Schmitz and the many other people with whom I had interesting discussions during the planning phase of this project.

Many thanks also to Ek Del Val, Ryan Keane, Josie Harral and Emma Pilgrim from Mick’s group at Silwood Park, and to Maria G.Alvarez (Montpellier, France); Brian Pickett for filling the pots with me; Maria Magalhaes, Bish Das and Doris Leung for always being there for me at Silwood, and of course to all the other Silwoodians with whom I have shared such unforgettable times; Gary Brown for his support and encouragement to come to the UK, and Katrin Meyer – who has introduced me to the Silwood community.

Part of this project was supported by the German National Merit Foundation.

Declaration

Hiermit versichere ich, dass ich die vorliegende Diplomarbeit selbständig verfasst und keine anderen als die angegebenen Quellen und Hilfsmittel verwendet habe.

Abbildung in dieser Leseprobe nicht enthalten

„Nowadays we live in a very explosive world,

and while we may not know

where or when the next outburst will be,

we might hope to find ways of stopping it

or at any rate damping down its force.“

Charles S.Elton 1958, p.15

Preface

My first encounter with an invasive plant was during my community service near the beautiful Lake Starnberg in Southern Germany, in summer 1996. We were equipped with spates, raincoats, and brushcutters with circular saw blades – fighting against dense stands of tree-like Giant Hogweed (Heracleum mantegazzianum Somme. & Lev.) introduced from the Caucasus, which is probably one of the best-known alien plants in Europe. Exposure to the sap of this plant causes phytophotodermatitis (Lagey et al. 1995) and results in swelling, blisters and eruptions of affected sites.

Some people may think that biological invasions are nothing to worry about – but in many areas of the world, invasive organisms are already causing serious ecological and economical problems, and there is increasing evidence that biological invasions can in fact be regarded as one of the most important components of human-caused global change.

Yet, surprisingly little is known about the biology of some invasive species. Without this knowledge, however, and especially without experimental support from carefully designed and replicated experiments, assumptions on the possible impact of a specific invading organism may be misleading.

In this study, I will focus on a species which has only recently arrived in Europe, the Narrow-leaved Ragwort (Senecio inaequidens DC). An up-to-date summary of current knowlegde about the biology of this species will be provided, and I hope that at least some of the experimental results will be of help for further studies on this invasive plant.

List of Figures

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

List of Plates

Plate 1

Plate 2

Plate 3

Plate 4

Plate 5

Plate 6

Plate 7

Plate 8

Plate 9

Plate 10

1 Introduction

1.1 Preliminary remarks

The nomenclature of species in this work is according to Wisskirchen & Haeupler 1998 and Schaefer 2000.

As far as possible, characters of non-English languages (e.g. German umlauts, French accented letters) have been overtaken properly, but especially within the reference list there were restrictions due to the software used. Footnotes will be presented on the bottom of each page or on one of the following pages.

1.2 Why study biological invasions?

1.2.1 The borderless world

There can be no doubt that the human population has grown to a dangerous size (e.g. UN 2001, Bernstein 2001) and there is still no evidence for a decline in the near future (but see Lutz et al. 2001 and Smil 1999 for long-term predictions). And with this growth in population size, there is more need for resources, be it in terms of spatial, biotic or abiotic resources. Consequently, as man is altering the resource distribution and composition on earth, there is growing evidence for large environmental changes to take place, namely changes in climate, biotic productivity, water resources, atmospheric chemistry, and ecological systems - commonly referred to as ‚global change’ (see Grime 1997 and Grime 1997; Vitousek, Mooney et al. 1997 for a general overview).

Since the end of the 19th century, major inventions like the invention of the automobile or the aircraft have led to an exponential increase in per-capita mobility of humans. Schafer & Victor 1999 estimate the worldwide traffic volume to have risen from 6 - 1012 person kilometres in 1960, to 28 - 1012 pkm in 1990. Generally speaking, it can be stated that, for all regions of the world, there is a linear correlation between per-capita gross domestic product and per-capita traffic volume. This means that the more industrialized a nation becomes, the more demand for high-speed transportations there will be.

This ever increasing far-distance transport of people and their goods, however, has lead to a breakdown of biogeographic barriers – different types of organisms are brought into regions where they would normally never occur naturally.

Organisms that have been introduced to a given area outside their native range by means of human-mediated transportation, are commonly called ‘alien’ or ‘exotic’ organisms (Keane & Crawley 2002; Richardson et al. 2000; Sax & Brown 2000); in the case of plants, it can be stated that

(1) if they fullfill the ‘invasion criterion’ (Crawley 1997a), they are called ‘naturalized’ alien plant species (see chapter 1.2.3.1 ); and
(2) if they are able to reproduce in natural habitats, they are called ‘invasive’ alien plant species¹.

As Mooney 1999 points out, the resulting biological “scramble of species” may have “serious consequences for the future course of evolution and the number of species that will populate the earth”. The major consequences discussed in context with biological invasions are for example²:

- altered species composition and structure of existing communities (e.g. Kowarik 1990)
- altered ecosystem-level rates of ressource supply, altered trophic structure, and altered disturbance regime of the invaded areas (D'Antonio & Vitousek 1992)
- reduced abundance and sometimes even total displacement of native species – e.g. by means of competitive suppression, altered disease incidence, trophic interactions, or changes in abiotic conditions (e.g. Tilman & Lehman 2001)
- loss of biodiversity (Vitousek, D'Antonio et al. 1997, cited in Keane & Crawley 2002)
- economical consequences, especially when invasive organisms affect agricultural ecosystems (e.g. Barbier 2001; Perrings et al. 2000)

Yet, surprisingly little is known about which organisms will become successful invaders, where and when invasions will occur, and what effects they may have. It seems that with the increase of speed in our global transportation systems, we have lost control over the consequences, and the “bombardment of every country by foreign species” (Elton 1958) will grow more and more intense. Thus, it is necessary to find quick and reliable ways to spot, predict and manage biological invasions.

1.2.2 Biological invasions: concepts and definitions

Every biological species originates in a single location and expands its range through migration or dispersal³. This range expansion is called biological invasion – one species invades an area in which it previously did not occur⁴. Climatic, geographic or topographic barriers as well as biotic and abiotic factors limit the spatial and temporal distribution of a species, and may hence form barriers that prevent biological invasions.

Biological invasion per se is a process as fundamental for biological systems as Brownian movement⁵ for quantum mechanics. As Crawley 1997a, p. 617 ff. points out, all species that are persistent in a specific habitat pass the invasion criterion:

Abbildung in dieser Leseprobe nicht enthalten

This means that all species possess the ability to increase their population size when their population densities (N) are low (Crawley 1997a); hence, every attempt to search for specific traits of “invasive” and “non-invasive” species will lead to wrong conclusions. However, it is of course possible to make assumptions on the “degree of invasiveness” of a certain species under given environmental conditions, but this will always be a function of abiotic and biotic conditions, time, and interactions between these factors.

Despite all these theoretical considerations, there can be no doubt that human-mediated long-distance dispersal of species has led to an increase in dispersal and migration rates in organisms, and that this increase will result in changes in biological communities. Biological invasions are “a significant component of human-caused global change” (Dukes & Mooney 1999; Lövei 1997; Vitousek, D'Antonio et al. 1997), and it is therefore necessary to clearly define terms and definitions used in this context.

- Alien organisms have been introduced to a given area outside their native range by means of human-mediated transportation (Keane & Crawley 2002). For plants, a dispersal distance of >100 km is the minimum requirement for this term to be applied (Richardson et al. 2000)
- Naturalized organisms are alien organisms that reproduce consistently, and sustain populations over many life cycles without direct intervention by humans (Richardson et al. 2000)
- Invasive organisms shall be defined here⁶ as naturalized organisms that produce reproductive offspring usually at considerable distances from the parent organisms (Rejmanek 2000; Richardson et al. 2000); for plants reproducing by means of seeds or other propagules, a spread of >100m in less than 50 years shall be used as a criterion (Richardson et al. 2000)

1.2.3 Invasive plants

1.2.3.1 General overview

The global extent of invasions by terrestrial plants is difficult to estimate (see Lonsdale 1999 for an overview), and it is even difficult to generalize on taxonomic patterns; however, it can be stated that the families Amaranthaceae, Brassicaceae, Chenopodiaceae, Fabaceae, Hydrocharitaceae, Papaveraceae, Poaceae and Polygonaceae show exceptionally high numbers of invasive species (Rejmanek 2000). The total number of introduced species is highest for the floras of Australia and New Zealand⁷ (Heywood 1989). Some commonly accepted generalizations (closely following Rejmanek 2000) on invasive plants are

- the probability of invasion success increases with initial population size and with the number of introduction attempts
- the spread of many alien plant species is highly dependent on human activity (e.g. dispersal by human-mediated transportation)
- the increased abundance of invasive alien plant species outside their native range is due to a decrease in regulation by herbivores and other natural enemies (enemy release hypothesis, Keane & Crawley 2002; Sax & Brown 2000)
- plant communities with high species diversity are more resistant to invasion (diversity resistance hypothesis; Kennedy et al. 2002)
- small genome size, small seed size, high leaf area ratio and high relative growth rate of seedlings may be “an ultimate determinant of plant invasiveness” (Rejmanek 2000)
- the size of the native geographical range⁸ of an herbaceous species may predict invasiveness (Keane, personal communication)
- alien species belonging to exotic genera are more likely to be invasive than those with native congeners
- plant species depending on non-specific mutualisms are more likely to overcome abiotic and biotic barriers in new enviroments
- there are often lag phases between introduction and first spontaneous spread of an alien plant (Kowarik 1995b)

The different stages of biological invasions (survival in transport, establishment in areas outside the native range, lag period, and spread in non-native areas), as well as further characteristics have been summarized by various authors, e.g. Sakai et al. 2001. Abbott 1992 mentions that, in case of plant invasions, interspecific hybridizations between native and exotic species may play a major role in the evolution of new plant taxa.

1.2.3.2 Important invasive alien plants in Europe

In the Central European context, it has been common to classify alien plants as “neophytes” and “archaeophytes”, depending on their date of first spontaneous appearance (after or before 1492, respectively). There are more detailed classification systems, which are fully covered in Kowarik 1999.

Up to now, about 12,000 ferns and vascular plants have been imported to Central Europe, less than 5% of which having established permanently (Kowarik 1999). For Britain, Crawley 1997a states about 20,000 introduced vascular plants. Neophytes regarded as exceptionally important are for example⁹ (largely based on Crawley 1997a and Kowarik 1995a):

- Heracleum mantegazzianum Sommier & Levier (Apiaceae)
- Impatiens glandulifera Royle (Balsaminaceae)
- Solidago canadensis L. (Asteraceae)
- Helianthus tuberosus L. (Asteraceae)
- Reynoutria japonica Houtt. (Polygonaceae) (Böhmer et al. 2001)
- Spartina anglica C.E.Hubbard (Poaceae)
- Senecio inaequidens DC. (Asteraceae), the species dealt with in this thesis (Plate 1 a, 7 b)

1.3 Herbivory and plant competition

1.3.1 Herbivory

Herbivory is an antagonistic relationship (+/-) between plants and animals (Strauss & Zangerl 2002). Herbivores are animals that feed exclusively on living plant tissues (Crawley 1983), e.g. grazers, phytophages, granivores, frugivores or sapsuckers.

Herbivores can be classified according to their mode of feeding (e.g. sucking, stripping, mining), the tissues aten (granivores, frugivores etc.), the host plant spectrum (specialists, generalists – see below) as well as their phylogenetic position (invertebrate, vertebrate, insect herbivores etc) - see Strauss & Zangerl 2002 for details.

The effects of herbivores on the performance and population dynamics of a plant can vary greatly. Sometimes whole plants are killed immediately (e.g. seed and seedling herbivory). Other herbivores may not directly cause plant mortality and instead even increase plant fecundity, as is the case for plants that are able to overcompensate for herbivory (Crawley 1997b). It needs to be stated that even low herbivore densities may have significant effects on plant performance and fecundity, if the dispersal rate and searching efficiency of the animal are high (Harper 1977, cited in Whittaker 1978). It is furthermore important to note that several herbivores may also be involved in pathogen transmission, e.g. aphids that may transmit plant viruses.

For the purpose of this study, the following definitions regarding the degree of host specificity in herbivores shall be used:

- generalist herbivores shall be defined as herbivores that feed on many different plant species
- specialist (oligophagous) herbivores feed on only a few plant species, usually of the same genus
- monophagous species are those that feed exclusively on one single host plant species.

However, even within one plant species, host races may alter the performance of monophagous herbivores (Wink & Legal 2001).

1.3.2 Plant competition

Plant competition is an antagonistic relationship (-/-) between individual plants of one or different plant species. Competition generally changes or stabilizes the composition of mixtures (Silvertown & Charlesworth 2001).

There are several ways to define plant competition:

(1) Mechanistic definition: neighbouring plant species compete for the same resource (e.g. water or light)
(2) Demographic definition: the net reproductive rate of one or both species is lowered

Competition can be intra- or interspecific, and in some cases it may furthermore be necessary to differentiate between above- and belowground competition (McPhee & Aarssen 2001). In this study, the focus shall be on aboveground competition.

As stated by Watkinson 1997, when trying to measure the effects of plant competition, it is important not only to measure final biomass or growth, but also to take into account the survival and fecundity of plants in mixtures. Another important point to notice is that density-dependent effects can be important in regulating the number and sizes of individuals in plant populations.

In the study presented here, competition levels (0 and 1) will be varied by total removal of the vegetation cover; in a strict sense, this could be regarded a disturbance treatment, but it shall be referrerd to as ‘competition’ treatment from this point onward, as long as not stated otherwise.

1.4 The special case of Senecio inaequidens DC

1.4.1 The main reasons for focussing on this species

Senecio inaequidens offers several unique opportunities as a study object:

- its actual rate of spread in Europe is exceptionally high¹⁰
- it has close relatives in Europe, and native specialist herbivores might well accept this species, despite the predictions of the ’enemy release hypothesis’¹¹
- it contains toxic compounds¹² and might become a ‘problem plant¹³ ’ if starting to invade pasturelands and agricultural fields in the near future
- very little is known about its biology
- there are no experimental data on wether or not S.inaequidens can invade grassland ecosystems
- without these data, no decisions can be made on how to manage ecological or economical impacts that the invasion of S.inaequidens might have

1.4.2 Phylogenetic position

The genus Senecio L. (Asteraceae) comprises about 1000-3000 species and has been subdivided into approximately 150 sections (Pelser et al. 2002). According to these authors, the genus is presumably “paraphyletic or even polyphyletic”. About 180 Senecio species can be found in South Africa (Hilliard 1977).¹⁴

Senecio inaequidens DC belongs to the section Fruticulosi DC and the tribe Senecioneae within the family Asteraceae (Clapham et al. 1987). The type specimen of Senecio inaequidens DC was collected in South Africa by Drège (Nr.5879, Herbarium G-DC; Hilliard 1977) and described as S.inaequidens by de Candolle 1837.
Synonyms are:

- S.lautus auct. non Soland ex Willdenow (Stace 1997)
- Senecio burchellii DC. p.p. (Hilliard 1977)

Some of the most often used common names include:

- narrow-leaved ragwort, canary Weed, Molteno disease Senecio, Burchell Senecio (in English-speaking countries; Bromilow 1995)
- séneçon du Cap (France; Troussel 1998)
- Schmalblättriges Greiskraut, Schmalblättriges Kreuzkraut, Gleiskraut, Ungeradzähniges Greiskraut (German-speaking countries)
- senecione sudafricano (Italy)
- “la escondilla” (Colombia; Herrera 2000)
- Geelopslag (South Africa; Bromilow 1995)

In Europe, S.inaequidens has presumably been confused with other Senecio species (according to Böhmer et al. 2001, Werner et al. 1991, and Kuhbier 1977), such as:

- S.lautus (Soland ex Forster) A.Richard
- S.lautus Forster f. ex Willdenow
- S.harveianus Mac Owan
- S.vimineus Harvey non DC.
- S.reclinatus L.f.
- S.linifolius L.
- S.paniculatus Berg.
- S.douglasii DC.
- S.burchellii DC.
- S.carnulentis DC., and
- S.fasciculatus ssp. minor Schlecht.

Several authors, including Radford et al. 2000, Radford & Cousens 2000, and Sindel et al. 1998 have discussed the phylogenetic of this species and especially its relationship to Senecio madagascariensis Poir that has been introduced to Australia from South Africa. A molecular phylogeny of Senecio sect. Jacobaea (Mill) Dumort including some data on the taxonomic position of Senecio inaequidens has recently been published by Pelser et al. 2002. Unfortunately, these authors included only S.inaequidens out of sect. Fruticulosi into their cladogram for Senecioneae, and further data on members of the Fruticulosi section would be needed to resolve the phylogenetic relationships of S.inaequidens.

In a recent yet unpublished phylogenetic analysis, W. Kadereit¹⁵ places S.inaequidens together with S.cryphiactis and S.abruptus into a South African clade, including the Mediterranean Senecio malacitanus. This clade again is included into a larger clade with Mediterranean and North American Senecio species. On the other hand, Radford et al. 2000 conclude from isozyme and morphological studies (including scanning electron micrograph analyses of achene surface morphology) that S.inaequidens belongs to a complex of closely related species from South Africa and Madagascar, including S. madagascariensis Poir, S.skirrhodon DC, S.burchellii DC and S.pellucidus DC. Stefan Neser¹⁶ names Senecio harveianus MacOwan, S.burchellii DC, S.skirrhodon DC, S.inaequidens DC and S.madagascariensis Poir as belonging to a complex that has been called the “S.madagascariensis / S.inaequidens complex” by Scott et al. 1998.

1.4.3 Country of origin and natural habitats

Senecio inaequidens is native¹⁷ to South Africa. In the following section, data on the distribution and natural habitats of this species and closely related members of the “S.madagascariensis complex” will be given, according to Adolphi 1997, Hilliard 1977 and personal communication with Dr.S.Neser¹⁸ and O.Bossdorf¹⁹. It is likely that hybridizations between Senecio inaequidens and other members of this complex have occurred, and many of the data given below may have to be revised and confirmed by additional observations, including biochemical traits and DNA fingerprinting.

The natural habitats in South Africa lie in the high-altitude so-called ‘Highveld’ areas (c. 1400 m – 2850 m above sea level, Hilliard 1977) in Freestate, Lesotho, Natal, Gauteng, as well as in the Northwest Province, Mpumalanga and the Northern Province of South Africa. Hilliard 1977 describes the original habitats of S.inaequidens as follows:

“Its natural habitat in Natal is among rock outcrops on steep, moist, grassy mountain slopes and along rocky watercourses, but it often becomes a weed along roadsides or firebreaks or in trampled or otherwise disturbed areas.”

Adolphi 1997 has found S.inaequidens on roadsides in the Featherbed Peninsula near Knysa, and in fynbos vegetation on the eastern side of the False Bay in South Africa, more or less close to the sea. Bromilow 1995 reports on weed occurrences of S.inaequidens in South Africa in wheat lands and other crop fields, gardens, roadsides and waste places, “mainly in the Cape especially in the Ceres, Middelburg and Prieska areas”.

According to Neser²⁰, herbarium records from the National Botanic Institute (South Africa) indicate that

- S. inaequidens DC and the closely related S.burchellii DC have a rather similar distribution “from the South West Cape Province diagonally across to near the North of the country, with a few records from Southern Namibia and adjacent areas”;
- S.burchellii was “relatively often” collected in the Western Cape area in the vicinity of Cape Town, whereas S.inaequidens records are “almost absent” in that part;
- S. madagascariensis POIR has been collected mainly near and along the Southeast coast from the Southern Cape area through KwaZulu-Natal with a few inland occurrences in Mpumalanga Province, in Gauteng, and also in the North of Northern Province of South Africa;
- S.skirrhodon is recorded as occurring only locally on the coast of the Eastern Cape and Mpumalanga Provinces;
- S.inaequidens, S.burchellii and S.madagascariensis have also been collected in Pretoria.

1.4.4 Invasion history

1.4.4.1 World (excluding Eurasia)

Due to the still unresolved phylogenetic position of Senecio inaequidens – and especially its unclear relationships with S.lautus (Soland ex Forster) A.Richard and S.madagascariensis Poir. (see section 1.4.2, p. 6) – there are still no reliable data on the distribution and the timing of arrival of S.inaequidens in areas outside Eurasia.

However, there is growing evidence that populations of S.inaequidens might already have established in parts of South America, Australia and New Zealand. Kuhbier 1977 and Stengl 1982 report on occurrences of S.inaequidens in Argentina and South Australia. Further support for occurrences in these countries comes from internet sources²¹. Asmus 1988 (cited in Wieners 1994) furthermore reports on possible occurrences in New Zealand.

In Colombia, Senecio inaequidens already seems to occur in agricultural ecosystems. Herrera 2000 and 2002 (unpublished) reports on successful biological control of Senecio inaequidens in Colombia, using the pyralid moth Homoeosoma oconequensis DYAR (Lepidoptera: Pyralidae). Records of Senecio inaequidens from Colombia are also supported by data presented by Robinson et al. 2002. In summary, there are indications that S.inaequidens or closely related species out of the ‘ S.madagascariensis complex’ already occur in

- (South) Australia
- New Zealand
- Argentina
- Colombia

Other parts of South America, especially in the Andean region, as well as regions with mediterranean or temperate climate, are likely to be invaded in the future.

1.4.4.2 Eurasia

S.inaequidens has been introduced to Europe together with wool transports from South Africa (Schmitz & Werner 2000). The earliest documented occurrences have been summarized by Ernst 1998 and Kuhbier 1977; the first European herbarium specimen have been collected around 1896 in Bremen (North Germany) near wool washeries. At all locations S.inaequidens has first been recorded from the vicinity of wool washeries or other wool-processing factories.

In the initial phase of spread, there have been at least five dispersal centres located at Mazamet (Southern France), Calais (Northern France), Verona (Italy), Liege (Belgium) and Bremen (Germany) - see Ernst 1998,Werner et al. 1991 and Kuhbier 1977 for details. Since the 1970s , S.inaequidens has been spreading quickly throughout parts of Central, Western and Southern Europe. The rapid spread of S.inaequidens has been explained by its ability to migrate along motorways and railway tracks, and this process has been well documented by Griese 1998 and 1996.

A detailed summary describing the whole process of invasion of S.inaequidens in Central Europe (with focus on Germany) has been published by Böhmer et al. 2001.

At the moment (August 2002), Senecio inaequidens has been recorded from

- Spain (especially Catalonia, see for example Vicens 1996)
- France (“Montpellier”, Troussel 1998; “Mazamet”, Gaida & Schneider-Gaida 1999; “dunes near Calais”, De Langhe et al. 1973, cited in Kuhbier 1977)
- Italy (“Verona”, Pignatti 1982, cited in Brandes 1999; “widespread in the Northeast”, Bicchi et al. 1985; “Lago di Garda”, collected by Stengl 1982; at high altitudes > 1000m (Kuhbier 1977))
- Denmark (Henker 1996, cited in Schmitz & Werner 2000)
- Belgium (cited in Ernst 1998)
- The Netherlands (Ernst 1998)
- Luxembourg (Colling & Reichling 1996, cited in Schmitz & Werner 2000)
- Switzerland (Schmitz & Werner 2000)
- Austria (Polatschek 1984, cited in Schmitz & Werner 2000)
- Poland (“Kattowice”, Ernst 1997, unpublished, cited in Ernst 1998)
- Norway (Often 1997, cited in Schmitz & Werner 2000)
- Finland²²
- Great Britain (“Dover”, “R[iver] Tweed”, “S[outh] Scotland”, Clapham et al. 1987).

1.4.4.3 Germany

A large amount of data has been published describing the history of introduction and the dynamics of spread of S.inaequidens in Germany. Detailed summaries can be found in Radkowitsch 1997 and Werner 2000. The initial phase of spread seems to have started from two dispersal centres, one in Northern Germany (Bremen; Kuhbier 1977) and one in Western Germany (Aachen, Cologne; Radkowitsch 1997). According to Werner et al. 1991, the city of Hannover has recently become a major new dispersal centre for S.inaequidens. Clearly, North Rhine Westphalia and Lower Saxony show the highest number of records per square. There seems to be a strong correlation between the density of major transport pathways (motorways, railway tracks etc), and the current distribution of this species.

For Germany, the easternmost occurrences are in Eastern Saxony (Niederschlesischer Oberlausitzkreis, near the river Lausitzer Neiße, Kartenblatt (TK25): 4655 Rothenburg²³ (Oberlausitz), and on the island of Rügen (Mecklenburg-Vorpommern, Kartenblatt (TK25): 1447 Saßnitz²⁴ ; Litterski & Berg 2000).

1.4.4.4 United Kingdom

For the UK, occurrences of S.inaequidens have been recorded from Scotland (Edinburgh), and also from some counties in the South, especially Kent.

In Berkshire, where the experiments for this study were performed, there are up to now two locations at which S.inaequidens has been found²⁵:

a) East Berkshire: as an escape from the Botanic Gardens in Whiteknights Park in 1988 (Herbarium RNG in Reading).
b) West Berkshire: at Abingdon in 1917 (Plant Sciences Herbarium OXF, Oxford).

1.4.5 Biological characteristics

1.4.5.1 Morphology and life form

According to Clapham et al. 1987 and Hilliard 1977, members of the genus Senecio L. may be characterised by the naked, epaleous receptacle, the smooth pappus consisting only of simple hairs, and the uniseriate involucral bracts. The ray florets are female, whereas the disc florets are hermaphrodite.

a) general morphology

Senecio inaequidens is a subglabrous, erect or often spreading chamaephytic perennial of up to 80-100 cm height²⁶, often forming stems branching from a more or less woody base (in contrast, S.madagascariensis forms stems that are “often simple below”, Hilliard 1977). The up to 50 (own observations) 1st order branches are often branching again several times, making the plants appear bushy. Basal branches with contact to the soil surface are capable of producing adventitious routs; branches or parts of the stem that have been bent over and/or wounded are also capable of forming adventitious roots.

b) leaf morphology

The leaves produced under high light intensities and moderate to low soil water content are approximately 2.5 cm long and 1-2(-7) mm wide (Clapham et al. 1987, Adolphi 1997), sessile, (mostly) linear with more or less entire or slightly irregular denticulate margins and with an auriculate, half-amplexicaulous base.

Leaf morphology is highly variable, presumably depending on light intensity, soil water potential, relative humidity and plant age (own observations). In low light intensity, especially under altered UV regime and greenhouse conditions, individuals of the same ecotype differ significantly in leaf morphology from the description given above; they produce leaves that are often up to 15 mm broad and more than 15 cm long (Plate 10 a), and even serratulate forms with teeth of about 10 mm length can be found (own observations; see Plate 7 b). In the leaf axils, S.inaequidens forms tufts of lanceolate axillary leaves that may grow out to form additional axillary shoots. The leaves are amphistomatic (Wieners 1994).

c) capitula morphology and structure

The flower heads (Plate 1 a) measure 2-2.5 cm in diameter (Clapham et al. 1987). There are (7-)13(-15) clear yellow (“canary yellow”, Hilliard 1977) ray-florets, each 5-8(-10)mm long (Stace 1997), approximately 70 clear yellow disk-florets (de Candolle 1837) and 10-20 short outer involucral bracts about (4-)5(-7) mm long (Hilliard 1977). The involucral bracts are keeled, with 1-3 nerves (Hilliard 1977) and fimbriate margins (Clapham et al. 1987). S.inaequidens shows only few calyculus bracts, which are often dark-tipped, non-overlapping and “much shorter than the involucre” (Hilliard 1977). The involucre itself is campanulate in shape.

d) achene morphology

The achenes are cylindrical, 2-2.5 mm long (Hilliard 1977) and pubescent between the ribs (Hilliard 1977, de Candolle 1837). Achene surface micromorphology is used for classification .

e) chromosome number

The base chromosome number is 2n=40 (Clapham et al. 1987).

1.4.5.2 Reproductive biology

a) pollination

S.inaequidens is mostly pollinated by

- Hymenoptera (Apidae, especially Apis mellifica and Bombus spec.) and
- Diptera (especially Syrphidae; Plate 1 a).

Other flower-visiting (but not necessarily pollinating) insects include

- Coleoptera (presumably pollen-feeding),
- Heteroptera (mostly achene-feeding or predatory) and
- Lepidoptera (e.g. Pieris brassicae L., photograph in Wieners 1994).

[...]

---

¹ M.J.Crawley, personal communication, August 2002.

² see e.g. Sakai et al. 2001 for a detailed review

³ Local short-distance movements shall not be considered here. See Cain et al. 2000 for aspects of long-distance dispersal.

⁴ In a strict sense, terms like “invasive species” or “alien species” are value-laden and should be replaced by more rational terms (Trudgill 2001); however, throughout the course of this thesis, they shall be used because they are commonly accepted terms in invasion biology.

⁵ as described, for example, in Einstein 1985

⁶ the terms ‘invasive’ (and ‘invasive alien’) shall be used here only for alien (exotic) organisms that pass the invasion criterion in given habitats outside their native range

⁷ smaller oceanic islands like Hawaii are a special case and shall not be considered here

⁸ see Sax & Brown 2000 for more characteristics of plant species with large geographical range sizes

⁹ woody species not taken into account here

¹⁰ Böhmer et al. 2001

¹¹ Keane & Crawley 2002

¹² see Biological characteristics of Senecio inaequidens

¹³ sensu Crawley 1997a

¹⁴ the term ‘phylogenetic’ is used here rather than ‘systematic’ or ‘taxonomic’; D.L.J.Quicke, Silwood Park, personal communication.

¹⁵ E-mail correspondence, 09.05.2002

¹⁶ e-mail correspondence, 31.01.2002, Plant Protection Research Institute, National Department of Agriculture, Pretoria, ZA

¹⁷ wether or not it can be regarded as being endemic to the Cape floral kingdom needs to be discussed in the future

¹⁸ E-mail correspondence, 31.01.2002, Plant Protection Research Institute, National Department of Agriculture, Pretoria, ZA

¹⁹ E-mail correspondence, 4.4.2002, UFZ Centre for Environmental Research Leipzig-Halle, Halle, Germany

²⁰ E-mail correspondence, 31.01.2002, Plant Protection Research Institute, National Department of Agriculture, Pretoria, ZA

²¹ http://www.school.za/projects/ptw/description.htm

²² http://www.helsinki.fi/lehdet/uh/299b.htm

²³ Datenbank Gefäßpflanzen am Bundesamt für Naturschutz und Zentralstelle für Phytodiversität, Germany, distribution map from 12/1999; “TK” means topographical map with scale 1:25,000

²⁴ see 23

²⁵ M.Crawley, e-mail correspondence, 3rd September 2002

²⁶ Under greenhouse conditions, a maximum height of 210 has been recorded (own observations)