Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of clonal structure and mating patterns in Maianthemum dilatatum Wilson, Amy 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2004-0135.pdf [ 5.13MB ]
JSON: 831-1.0075049.json
JSON-LD: 831-1.0075049-ld.json
RDF/XML (Pretty): 831-1.0075049-rdf.xml
RDF/JSON: 831-1.0075049-rdf.json
Turtle: 831-1.0075049-turtle.txt
N-Triples: 831-1.0075049-rdf-ntriples.txt
Original Record: 831-1.0075049-source.json
Full Text

Full Text

Characterization of clonal structure and mating patterns in Maianthemum dilatatum by Amy Wilson B.Sc. (Hons.), The University of Calgary, 1999 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF FOREST SCIENCES We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA February 2004 © Amy Wilson Library Authorization In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Name of Author (please print) Date (dd/mm/yyyy) Title of Thesis: Degree: j f lQG^ Year: 20^ Department of fnfcbST iSftr^S The University of British Columbia Vancouver, BC Canada ABSTRACT Maianthemum dilatatum is a liliaceous, rhizomatous herb, which adopts a guerrilla mode of lateral spread. The central focus of this study was to characterize the clonal structure of Maianthemum dilatatum using AFLP markers. A subpopulation was selected where 21 patches, covering a total area of 3236 m2 were mapped and sampled. Within these patches, 116 ramets were sampled and assigned to 74 distinct genets. Patches were genetically heterogeneous and moderately differentiated (0ST =0.291, p=0.001). Multivariate Mantel tests detected a genetic patch width of approximately 55-m, to which clonal growth contributed up to 25-m. This genetic patch was concluded to be the cumulative result of seed and pollen dispersal distances. Evidence of genet natality was present with the detection of five yearlings. Therefore, within M. dilatatum populations, clonality is a significant factor but the spatial structuring of genetic variation suggests that both low levels of restricted gene flow and repeated recruitment of genets occur. It has often been predicted that clonal plants will suffer increased geitonogamy (within-genet self pollination) as a genet increases in size. The prerequisites for this relationship to exist are: a high clonemate encounter probability and highly leptokurtic pollinator visitation, which have not been tested in a guerrilla morphology such as Maianthemum dilatatum. Clonemate encounter probability was measured at the ramet level, and was determined to account for 80% of all ten near-neighbours. Pollinators did preferentially forage among near-neighbours but they typically left the plots after visiting an average of only two floral ramets, which lowers the expected rate of geitonogamy based on genet arrangement alone. Therefore, the prerequisites for clonemate geitonogamy did exist in M. dilatatum, namely the high frequency of clonemate encounter and leptokurtic pollinator visitation patterns. Secondly, the contribution of geitonogamy to female reproductive success was estimated through hand pollinations and pollinator exclusion. Hand pollinations confirmed that M. dilatatum's self-incompatibility and pollen limited. An interaction was found between pollination and pollinator exclusion treatments (x2=92.399, pO.OOl), where excluding pollinators from outcrossed individuals increased fruit set. This supports the self-interference hypothesis, which is a predicted consequence of geitonogamy within self-incompatible plants. Emasculation did not have a significant effect on seed set (x2=l .0991, p=0.2945) which implies that ramet-level geitonogamy is not a significant contributor to self-interference. Therefore, in a guerrilla species such as Maianthemum dilatatum, clonemate geitonogamy is a factor constraining female reproductive success, but to a lesser degree than would be expected in a species with a phalanx morphology. In conclusion, clonal growth makes a significant contribution to the spatial structure and to mating patterns within M. dilatatum populations. Yet, repeated genet recruitment and genet intermingling allows for the high genotypic diversity, which reduces but does not eliminate the influence of geitonogamous pollination. ii TABLE OF CONTENTS Page Abstract ii Table of Contents iii List of Tables v List of Figures vi Acknowledgements vii CHAPTER 1 The Distribution and Consequences of Clonality- A review 1 1.1 Introduction 1 1.1.1 The genetic individual 2 1.1.2 The physical individual 2 1.2 Distribution of clonality within plant and algal groups 3 1.2.1 The evolutionary history of clonality 3 1.2.2 Clonality in plants and algae 3 1.2.3 Clonality in the Angiosperms 3 1.3 Consequences of clonality 4 1.3.1 Demographic consequences of clonality 4 1.3.2 Ecophysiology consequences of clonality 6 1.3.3 Conservation consequences of clonality 8 1.4 Conclusions 9 CHAPTER 2 The Clonal Structure of Maianthemum dilatatum 13 2.1 Introduction 13 2.1.1 Determining clonal structure 13 2.1.2 Genotypic and genetic variation within clonal populations 13 2.1.3 Spatial arrangement of genetic variation 14 2.1.4 Study objectives 15 2.2 Materials and Methods 15 2.2.1 Study Species 15 2.2.2 Study Site and Sampling 16 2.2.3 Estimates of sexual recruitment 16 2.2.4 Population level comparisons of genotypic and genetic variation 17 2.2.5 DNA extraction protocol for leaf tissue 17 2.2.6 AFLP protocol 17 2.2.7 Detection and scoring of AFLP 18 2.3 Data Analysis 19 2.3.1 Genotypic diversity 19 2.3.2 Genetic variation within the Pacific Spirit subpopulation 21 2.3.3 Spatial structure of genetic variation within populations 21 2.3.4 Population level comparisons of genotypic and genetic variation 22 2.4 Results 22 2.4.1 Genotypic diversity 22 2.4.2 Genetic variation within the Pacific Spirit subpopulation 23 2.4.3 Spatial structure of genetic variation within populations 23 2.4.4 Population level comparisons of genotypic and genetic variation 23 in Page 2.5 Discussion 24 2.5.1 AFLP reproducibility 24 2.5.2 Genotypic diversity and relatedness among genets 25 2.5.3 Genetic variation within the Pacific Spirit subpopulation 29 2.5.4 Spatial structure of genetic variation within populations 30 2.5.5 Population level comparisons of genotypic and genetic variation 32 2.6 Conclusions 33 CHAPTER 3 The Role of Clonal Growth in Pollination Patterns 40 3.1 Introduction 40 3.1.1 Pollination ecology 40 3.1.2 Clonality and pollination patterns 40 3.1.3 Reproductive biology of the study species 41 3.1.4 Study objectives 42 3.2 Materials and methods 42 3.2.1 Pollination experiments 42 3.2.2 Near-neighbour structure 43 3.2.3 Pollinator behaviour 43 3.2.4 Geitonogamy 44 3.2.5 DNA extraction protocol for leaf tissue 44 3.2.6 DNA extraction protocol for endosperm 44 3.2.7 AFLP protocol 45 3.3 Data analysis 45 3.3.1 Pollination experiments 45 3.3.2 Near neighbour structure 46 3.3.3 Pollinator behaviour 46 3.3.4 Geitonogamy 46 3.4 Results 47 3.4.1 Pollination experiments 47 3.4.2 Near neighbour spatial structure 47 3.4.3 Pollinator behaviour 48 3.4.4 Geitonogamy 49 3.5 Discussion 49 3.5.1 Pollination experiments 49 3.5.2 Near neighbour structure 50 3.5.3 Pollinator flight distances 51 3.5.4 Geitonogamy 53 3.6 Conclusions 55 Future Directions 60 Literature cited 61 i v LIST OF TABLES Page 1.1 Summary of the distribution of clonality within algal and plant divisions 10 1.2 Reported incidence of clonality in Angiosperm families present in BC. The family name has two superscripts the first being the clonal followed by the aclonal tendency. A superscript 1 designates confirmed presence of either state and 0 confirms the absence of either state while a ? designates that no information was available 11 2.1 Sequences of AFLP Primers used in this study 34 2.2 Patch Characteristics of Pacific Spirit subpopulation 34 2.3 Summary of the Lily of the Valley, patch-level AMOVA analysis on all seven multi-genet patches with significance levels of p=0.001 which was based on 999 permutations 34 3.1 Near-neighbour genetic structure within 1 -m diameter circular plot 57 3.2 Visitation behaviour of floral visitors in terms of flight distances (cm), number of ramets visited per 1-m plot, and near-neighbour transfer tendency given as weighted average ± SE 57 v LIST OF FIGURES Page 1.1 Proportionate representation of clonality among plant families in the fossil record. Data taken directly from Tiffney and Niklas (1985) 12 2.1 Frequency distribution of pair-wise Jaccard similarities of known siblings and clonemates 35 2.2 Decline of matrix incompatibility count with sequential genet removal 35 2.3 Frequency distribution of pair-wise relatedness coefficients of all sampled ramets within the Pacific Spirit subpopulation 36 2.4 Distribution of genets across patches within Pacific Spirit subpopulation. All genets are designated as a separate number where clonemates are given the same number. The positions where yearlings were found are marked with an® 37 2.5 Mantel correlation test between genetic and geographic distance all sampled ramets in the Pacific Spirit subpopulation (R2=0.1245, p=0.001) 38 2.6 Correlogram incorporating clonality, depicting r (correlation between genetic and geographical distance), 95% error bars about r with 95% CI given about the null hypothesis of r=0 38 2.7 Correlogram excluding clonality, depicting r (correlation between genetic and geographical distance), 95% error bars about r with 95% CI given about the null hypothesis of r=0 39 3.1 Seed set per individual ramet (weighted mean ± SE) after pollination and pollinator exclusion treatments (x2=92.399, pO.OOl) 58 3.2 Proportion of the ten nearest neighbours which were clonemates (self encounter) or different genets (compatible mates) 58 3.3 The flight tendencies within the ten nearest neighbour classes during a pollinator's flight transfer between ramets 59 3.4 The effect of emasculation on the number of seed set in open pollinated, unbagged plants (x2=l .0991, p=0.2945) 59 vi ACKNOWLEDGEMENTS I would like to thank my supervisor Dr. Bart van der Kamp for his guidance and for changing my research direction towards conservation genetics. I also benefited from the direction provided by my other committee members, Drs. Kermit Ritland, Rob Guy and Roy Turkington. For laboratory work I am indebted to Dr. Carol Ritland, who helped me through any technical and theoretical molecular problem and to Allyson Miscampbell. I also benefited from conservations with other lab members in the Ritland lab, in particular Yanik Berube. For field site access, I thank the Greater Vancouver Regional District, in particular Heidi Walsh and Jane Porter. I appreciatively acknowledge the support of the Natural Sciences and Engineering Research Council and the Sopron Alumni Association. For moral support I thank my husband Scott and my family. Additionally, I thank the Botany Department at the University of Calgary, for generous mentorship in the past. I dedicate this thesis but more the work I put into it, to the memory of my grandad, Jock (A.K) Anderson who passed away on January 29th, 2004. vii Chapter 1 The Distribution and Consequences of Clonality- A Review 1 1.1 Introduction Clonality is the replication of an individual by vegetative growth resulting in genetically identical, morphologically complete and potentially independent individuals (Harper 1977). Totipotency and modular organization render most plants capable of clonal growth, but it is only species that express it in natural, non-anthropogenic situations, which are considered to be clonal (van Groenendael et al. 1996). Clonal growth can be roughly categorized as either occurring as a part of a developmental program or facultatively. In developmental clonality, clonal reproduction occurs as part of development, which includes perennating organs such as rhizomes, stolons, bulbs, corms and adventitious buds. Facultative or opportunistic clonality would be regeneration from tissue that is not a specific organ of clonal growth, which is made possible by the persistent totipotency in plant tissues. Examples would be the regeneration from fragments (Urbanska 1989) and layering. The presence and extent of clonality has tremendous ecological, evolutionary and physiological consequences at both the level of the individual and the community. Clonality is widespread, occurring in plants, fungi and invertebrates with a tremendous variation in the mode and extent of expression. This chapter will first review pertinent definitions followed by a brief survey of the distribution of clonality within the plant kingdom along with some of the ecological consequences. The clonal habitat establishes what has often been referred to as a hierarchy of individuality, namely the genetic and the physical individual. These different levels differ in their demography, genetics and ecology so it is important that these differences are considered and that the terminology differentiating them is consistent. This not just an issue of semantics, inconsistent terminology results in miscommunication and misinterpretation of hypotheses, predications and data. Therefore, terminologies that will be adhered to throughout this thesis are described below. ' Information in portions of this chapter (1.1-1.2) have been submitted in a modified form as part of the requirements for a directed studies class. Permission for its inclusion was obtained from the instructor Dr. C. Ritland. 1 1.1.1. The genetic individual The genetic individual is termed the genet. The genet is the "genetic individual arising from a single zygote" (Harper and White 1974) constituting the "evolutionary individual" (Janzen 1977). Therefore, the term genet can be applied to both aclonal and clonal individuals. The difference is that in a clonal species, multiple individuals (ramets) can belong to a single genet, while in an aclonal species, the genet is only represented by a single individual. Another term used throughout this thesis is that of clonemates, which refers to ramets belonging to the same genet. Although not used widely, this term is very useful when discussing clonal structure. All clonemates necessarily share a common zygotic origin and theoretically an identical genome but in reality genetic fidelity is variable due to artifacts and somatic mutation. Artifactual bands arise due to contamination by exogenous DNA or due to the inhibition of PCR reactions by secondary compounds or sub-optimal reaction conditions. The frequency and role of somatic mutation as a source of variability and its adaptive significance within clonal species have attracted much discussion among evolutionary ecologists (Lushai et al. 2003). 1.1.2 The physical individual The easily discernable, physically distinct individual is called the ramet. Plants are modular organisms meaning that growth occurs by repeated addition of a base unit of growth, known as a module, and an example being a branch. Modules are interdependent but are not morphologically complete and thus could not survive independently. Like modules, ramets arise from meristems, but ramets are morphologically complete which allows them to gain autonomy. This independence criterion is what distinguishes a ramet from a module. A group of ramets which are physically connected are called a clonal fragment, wherein a single genet may exist as many variously sized clonal fragments. Since physiological integration among ramets requires the existence of both a physical and vascular connection, only ramets within a clonal fragment can participate in physiological integration. Studies of clonal species can use either a ramet or genet-based approach, it is only important that predictions and conclusions are scaled accordingly. For example it would be incorrect to study genet dynamics and to define all clonal fragments as separate genets, similarly one cannot use ramet-based data to test genet-based hypotheses. A more serious issue occurs when population viability censuses are completed with no mention of the underlying clonal structure. This is of particular importance given the crucial role that genetic factors have in the 2 accuracy of population viability analyses (Allendorf and Ryman 2002). Unfortunately, clonality is not discussed in the literature in proportion to its importance, which may be the result of the paucity of reports about its distribution. Therefore, one step towards an understanding of clonality is characterizing its distribution. 1.2 Distribution of clonality within plant and algal groups 1.2.1 The evolutionary history of clonality Paleobotanical accounts suggest clonality is the ancestral form of growth (Tiffney and Niklas 1985), supported by the fact that the progenitor of all vascular plants, Rhyniopsida was a clonal, rhizomatous plant (Stewart 1983, Mogie and Hutchings 1990). Throughout the fossil record (Fig. 1.1), there has been a decreasing prevalence of clonal growth, which was driven by the appearance of more derived non-clonal species coupled with the extinction of highly clonal species (Tiffney and Niklas 1985). In plants, patterns of clonality are determined by ontogeny, which is in turn a product of the phylogenetic history (Mogie and Hutchings 1990, van Groenendael et al. 1996). Mogie and Hutchings (1990) suggest that the distribution of a non-clonal habit was driven by evolutionary patterns in gametophyte dominance, embryo polarity and secondary growth. 1.2.2 Clonality in plants and algae Clonality is widespread within algal groups and non-vascular and seedless vascular plants. Taxa within these groups typically adopt the most archaic modes of clonal growth such as regeneration from fragmentation, budding and rhizomaty (Tiffney and Niklas 1985). Other common clonal organs are bulbils and gemmae. Bulbils and gemmae are vegetative propagules appearing in meristematic areas of stem tissue. Bulbils are axial buds, which have developed adventitious roots, detach and become independent ramets (Bell 1991); this is also known as pseudovivipary (Klimes et al. 1997). Typically associated with bryophytes, gemmae are propagules which differentiate from any meristematic tissue. Depending on the species, gemmae are dispersed in varying developmental stages from undifferentiated cell masses to small complete plantlets (Bold et al. 1987). Table 1.1 summarizes the clonal tendencies of three Algal divisions (Kingdom: Protoctista) and all of the non-vascular and vascular plant divisions excluding the Angiosperms (Anthophyta). 1.2.3 Clonality in the Angiosperms Angiosperms began their dominance in the early Cretaceous period with rapid diversification, culminating in varying degrees of clonality and a diverse array of clonal growth patterns. The Angiosperms are interesting evolutionarily because the retention or elimination of 3 clonality occurs as an ecological adaptation rather than as an ontogenetic obligation (Mogie and Hutchings 1990). Therefore, a greater diversity in clonal morphologies may arise, in order to exploit the various aspects of a clonal habit. The diversity of clonal morphologies within the Angiosperms is extensive and the reader is referred to Klimes el al. (1997) for a thorough treatment. Determining the adaptive significance of particular clonal growth forms is impeded by the inconsistent terminology and the sporadic reporting of clonal tendencies. To date, the only estimates of clonality within Angiosperms known to the author are 28% in dicots (Leakey 1981), >90% alpine and tundra habitats (Burrows 1990), >60% British Isles (Salisbury 1942, Abrahamson 1980) and 70% in temperate zones (van Groenendael and de Kroon 1990). Clonality is also an important life history trait in tropical ecosystems (Kinsman 1990, Sagers 1993). Available information regarding clonality for plant families in British Columbia was complied from Klimes et al. (1997), Hitchcock and Cronquist (1997) and Douglas et al. (1997) and given in Table 1.2. A single genus may contain both clonal and non-clonal members, so clonality was determined at the species level. For species whose clonality is not morphologically apparent, clonal tendencies would be inconclusive unless there was a specific description. The only rule of thumb is that annuals are almost never clonal (Richards 1986), so if species were reported as being annual this was taken as evidence for aclonality. As expected, information was limited but demonstrates that the majority of plant families within British Columbia have confirmed clonal members. 1.3 Consequences of clonality Clonality is a multi-function trait and as such has a large suite of potential consequences depending on the degree and mode of expression. Despite the large number of plant and invertebrate species which are clonal, and the availability of a volume of research, clonal life histories are essentially absent in even advanced demographic, ecological and physiological textbooks. This section will briefly summarize some of the demographic and ecological consequences of clonality. It is stressed that clonality is a multi-functional trait which is present in varying degrees among and within species such that any generalizations are limited. 1.3.1 Demographic consequences of clonality Demographic theory based on age-specific models is confounded in plants due to both plasticity and clonality (Harper 1980). Traditional demography is based on unitary species such that the census size is the true demographic and genetic size and where demographic parameters 4 can be predicted as a function of age. The demography of clonal plant species requires a different approach as the two types of individual, the ramet and the genet have demographically different properties (Caswell 1985). During the initial seedling recruitment stage, genet mortality should not differ from an aclonal plant, except if ramets competitively exclude seedlings (Abrahamson 1980). The differences in demography between clonal and aclonal genets only begin once the clonal genet becomes composed of more than one ramet. Harper (1980) refers to growth in clonal organisms as a demographic process where genet growth results in the 'birth' of ramets. Genet survivorship and fecundity is the product of survivorship and fecundity of these individual ramets such that genet-level dynamics are best described with size-structured models. Assuming sustained ramet production, genet survivorship and lifetime fecundity could continually increase (Harper and White 1974, Tanner 2001). Events of ramet mortality constitute partial genet mortality (Caswell 1985) such that correlated mortality among ramets reduces the survivorship probabilities of the genet as a whole. Generally, the more dispersed the ramets are, the greater the independence of mortality among ramets (Schmid 1990). The potential for extreme genet longevities is a significant consequence of clonal life histories, which would influence the nature of the evolutionary forces a clonal species would face (Silander 1985). Unlike the genets, ramets do have finite life spans such that they can be described using stage structured populations models. Stage structured models are most accurate for ramets given that plasticity results in age being a poor demographic predictor (Harper 1980, Caswell 1985). Birth rates for ramets have been shown to be both density dependent (Tanner 1999) and density independent (Schmid and Harper 1985). Closely integrated clonal morphologies such as the phalanx form, tend to have density dependent ramet birth rates such that self-interference is minimized and density dependent mortality is not significant (Schmid and Harper 1985). Conversely, guerilla species will tend to density independent ramet birth rates but density dependent death since ramet populations may overshoot carrying capacity (Lovett Doust 1981, Schmid and Harper 1985) Regardless of overall dynamics, new ramets have much lower mortality rates during early stages of recruitment than any genet recruiting from seed. This enhanced recruitment success is due to the omission of a vulnerable juvenile stage and the facilitation given by connected clonemates. Upon comparison with a fully established unitary individual, ramets should only differ in survivorship and fecundity as a function of physiological integration. 5 Conversely, Ehrlen and Lehtila (2002) in a meta-analysis of plant life histories concluded that ramets have a shorter life span than a unitary individual, which was attributed to the costs of clonal integration. Costs of clonal integration such as resource wastage on unproductive ramets and facilitated disease transfer would lead to non-independent mortality among ramets and 1.3.2 Ecophysiology consequences of clonality Clonal plants have an advantage of 1) increasing their spread and 2) the possibility of physiological integration. Mobility may allow a genet to forage locally perhaps by selectively placing ramets in resource-rich microsites (Schmid 1990). Increasing the internode length allows escape from poor habitats or competitors (Harper 1980) and slowing internode growth allows congregation in resource rich areas (Sutherland and Stillman 1988), or space consolidation and resistance to invasion (Schmid 1986). Physiological integration is the sharing of resources between physically connected ramets. Resources can be translocated bi-directionally as part of a division of labour process or when certain ramets have been stressed (Stuefer 1998). Division of labour occurs when ramets will maximize the collection of the most abundant resource at its immediate locale and translocate these resources to other ramets where resources differ in abundance. This is in contrast to a unitary plant which by necessity will maximize the most limiting resource. The advantage of any integration is that within a heterogeneous environment, ramets will sample different resource conditions such that ramets can complement limitations and integrate heterogeneity across the clonal fragment (Stuefer 1998). The duration and degree of cooperation of these connections varies both between and within species, and perhaps may occur simply as older ramets facilitating the recruitment of a new ramet or as permanent cooperation among ramets of varying age (Pitekla and Ashmun 1985). The extent of cooperation within a fragment is limited by vascular sectoriality such that physically connected ramets may not be connected physiologically (Marshall and Price 1997). Physiologically integrated ramets are termed integrated physiological units (IPUs) sensu Watson (1986) of which a clonal fragment can have several. Physiological integration has been demonstrated to facilitate pathogen escape (Wennstrom and Ericson 1992), herbivory recovery (Jonsdottir and Callaghan 1989) and to be involved in augmenting reproductive budgets of floral ramets within fragments (Ganger 1997). Physiological integration may seem to blur the independence of mortality of ramets and there have been demonstrations where systemic pathogens are transferred readily through these connections (Pitelka and Ashmun 1985). Kelly (1995) discusses the costs of integration and 6 suggests that integration has a concomitant requirement of disintegration when this non-independence of mortality threatens the clonal fragment. Strategies seem to be in place to preserve fragment survivorship at the expense of any single ramet (Schmid 1990). Disintegration has been documented in Solidago altissima where galls induced disintegration between infected and uninfected ramets (How et al. 1994). In fungal plant pathogen interactions studied in Lactuca sibirica (Wennstrom and Ericson 1992) and Carex arenaria (D'Hertefeldt and van der Putten 1998) integration fueled the increased dispersal of ramets such that genets essentially outran the pathogens. With respect to herbivore damage, Jonsdottir and Callaghan (1989) demonstrated in Carex bigelowii that a repeatedly defoliated ramet would only be subsidized twice before other ramets would stop providing resources to any ramet which was an excessive resource sink. Integrated ramets may also cooperate to maximize reproductive yields so that an integrated ramet would have an augmented budget allowing a higher fruit set than what would be possible without any facilitation (Ganger 1997). Clonal integration does not always increase competitive ability (Peltzer 2002) but is predicted to reduce the variance in competitive ability among genets (Gough et al. 2001). In terms of interspecific competition, the predicted role of clonality depends on the degree of lateral spread. Guerilla species disperse over greater scales of space, such that competitive encounters would be expected frequently. Phalanx species would tend to consolidate space and in doing so effectively resist invasion. In a competition experiment between a guerrilla (Prunella vulgaris), and a phalanx species (Bellis perennis), P. vulgaris could acquire space, but it could not retain it and thus was competitively inferior to the B. perennis at high densities (Schmid and Harper 1985). Therefore, the contribution of a clonal habit to the competitive ability of a species is largely a function of the tendency for lateral spread and integration. Pysek (1997) surveyed the role of the clonal habit on plant invasions, and found that as compared to aclonal species, clonal species only differed in the rate of spread and site occupation. Therefore, it is difficult to make generalizations on how being clonal changes the ecological role of a species, but it is clear that the impact of clonality would be expected to vary among ecological and successional communities (Pysek 1997, Gough et al. 2001). An interesting ecological aspect of clonal plants is the potential for self-interference. In clonally prolific species there is a high probability of encountering a clonemate. Self-interference can be mediated by physiological integration such that root competition can be avoided (Holzapfel and Alpert 2003), but in disconnected clonemates it would be more difficult 7 to discriminate a clonemate versus a conspecific. Falik et al. (2003) found in Pimm sativum that self-identification was mediated by physiological means but there was some evidence for allorecognition. Spreading clonal morphologies would result in situations where clonemates arising from different clonal fragments had never been connected, which would require non-physiological recognition mechanisms. Although the actual means of is poorly understood, root communication is a phenomenon that may have great importance for the avoidance of competition among clonemates (Falik et al. 2003). The ecological consequences of clonality are likely far-reaching but there is a need for additional empirical and comparative studies within natural populations. Interpreting the adaptiveness of clonal traits in particular habitats is complicated by the fact that similar clonal traits can be adaptive in very different situations (van Groenendael et al. 1996). Progress in understanding the adaptive importance of clonality in general and the diversity of clonal strategies would benefit greatly from a large-scale meta-analysis along with additional data from a broader range of habitats and species (van Groenendael et al. 1996). 1.3.3 Conservation consequences of clonality Commonality is often inferred from apparent abundance, but given that single genets can cover a considerable area, census-based methods are inappropriate for population viability studies. For example, a population may consist of 100 individuals but if these individuals are primarily clonemates then the prospects for population viability significantly decline. Generally, when genetic factors are incorporated into population viability analyses the estimated persistence time is more accurate and generally much shorter (Allendorf and Ryman 2002). Erikkson (1994) has shown that the ramifications of hindered genet recruitment may not manifest until decades later but sets that population onto a unavoidable trajectory towards extinction. The genetic future of a population is determined by the effective population size, so it is imperative that N e can be predicted accurately and efficiently. An accepted rule of thumb for unitary species is that N e is 10% of the total census size (Soule 1980). Therefore, if population assessments are ramet-based, both measures of genetic variation and population viability will be grossly overestimated. Additionally, clonal species may also need larger minimum habitat areas given that fewer genetic individuals are present on a per area basis than in unitary species. For endangered clonal species their decline seems heralded by reduced reproductive output and essentially clonality simply serves to delay the extinction (Johnson and Steiner 2000). 8 1.4 Conclusions Clonality is a ubiquitous trait in the plant kingdom yet descriptions of the extent of clonality within natural populations are still limited. Surprisingly, very few descriptions mention clonal tendency even in plants which are known from other sources to be highly clonal. One major, but obvious conservation implication for clonal organisms is that apparent abundance is an unreliable measure of population stability. Yet more subtle differences such as demographic patterns and physiology mean that responses to disturbances, minimum habitat area, gene flow scale, genetic rescue or population management cannot be predicted with theory developed for unitary organisms. Therefore, the management of clonal species needs to accommodate and exploit, clonal characteristics. Understanding the adaptive significance of clonality is of great interest to many researchers but this is limited by the absence of knowledge of clonality in most flora. Therefore, it is the goal of this thesis in Chapter two to describe the clonal structure in a native plant species and in Chapter three to attempt to measure the pollination consequences of clonal growth. 9 co p o 1 P ft o D . 13 cd c 2 CO rf o ft p o .S - f l rf cd o ui • — a -t-1 -3 , x ft _5 « • 15 * <u cn 2 Q cu -ti S « fc ° ^ > 5 I-I R c5 ^ o .3 J3 ° ^ co P co P O 'Lo '> c cd T3 I 13 13 c H—» 13 p o "o <4—l o p _o * s 3 r2 "C +-» CO "3 cu X! -•-» o & S a 13 rf o co _CJ 'o p <D 13 P CU rf T3 P crt 1) -a o < «.a ta o u a 1-g & a ft <» X S g § ° > I .3 ~ I •a ^ P " 5 n O ^ 3 ft o rf o CJ c h ^ -d c a o . 5 .2 o 0 3 -s '3 -cj C Cft " S CJ rf co ft u •J £Prf -a _ crt 55 Q • rf i - ^ 13 ( ^ i ft cd „ _ O 13 ' S 0 0 5 crt 'Lt fl =3 _M £ " 3 O -a O N C . rf P 1> — , s > .S B rf1 co •tJ O 1 .a « O >. CU co - O crt crt 11 s a s co CJ) ° © T d • 2 -S ft« a « ^ cm P a a 0 3 a; o o .2; CD _ 60 2 eg p X - 3 3 « ra N rf .2 i cS ^ M) S 13 2 < ffi X> 43 0 0 rf oo 3 0 S o r - - 1 o> cu O co CO " — ' rf. CO O cd (Dfl p o £ p Cd O H -3 <3 *—i CU JO T3 P CU - P o „ Crf a , —r cu cd > CO - r f cu cd P H t> .22 oo T3 CU u rf P cu - % .22 S 13 ^ 00 _p P Q cd P OH O P H 3 kH P cd C P JZ CO cu •3 3 I D CU £ o OH CU S-i O « 2 rf cd a a cd cu -rf oo co ^ 2"H rf 00 rf p 2 3 CJ o 1 s 2 - P a o CO J3 p p 8.2 ° ts CO rf 2 -3 DH A ^ a « ^ -P P .ti P Id u c o 00 cd co u 13 rf O cu ft^D CU •S3 '3 CO ^ rf t -P O O « cd P rf cd P o ft IS O rf DH C rf ^ 13 -rf cu V rf >, o .-p. ft 1* P g O ° 13 P cd 13 <Hrf o p p cd oa c £ o p cd - P 00 p o 13 cu > cu rf o cd P oo o oo T i <3^  o .rf° ffi rfH ^ 5 CO T H O CU cu P p 0:3 C+H - — ' rf >. ,o .-p. 1^—I rf co S CU P 3^ -2 3 O -rf co cu H< CO . 00 - P - • ce 3 m O N "2 C7\ ON P ON rf <H—I a c^ a> . Co ' P 13 ' CU £ O ft cu p 2 2 >^ -S cd DH p ^ " " rf CO -2 Q cd rf u -rf O 13 rf OO 2 cd ( X rf co" OO Jo °^ 3 co P - CU 5 CO 2 P CO w 0 0 O N . cd £ l ON t) rf ^ '3 Cd CO cu p o O N cd c Is cd N / o 13 'i3 00 -T3 ' P cd cd cd o rf w S P cd cd « cu u g 5 cu cd rffl CO ^ 3 P ^ O 13 "71 P u cd ° - P co 13 ? rf 2 3 oo CO > . rf cd 6 13 ft o 0 O 1 " a co cd cd oo ^ ^ 0 0 . .rf p o a a o o co cd o o o cd P 2 a -S op >- 2 CU CO cu *o >-i " P O P • £ ON <N ON o oo -S-drf--3 P ^ S 13 u U fl cd cd — rffl - P ° o i 2 n 2 . 2 ^ "3 3 o 3 3 t3 D . ft CU o o — ft ft g 13 P — «J •"S s O co rf " <" p S _0J to rf rf rf -C) rf o p _o *s rf 3 ft o ft cd a DH 13 CU p 13 P cd rf o co O / - v ft CN cu ON O N -CU cd  „fl U rf rf rf cd cd x ^ B u O N ^ S O N T3 - P rf cu (72 1 / 3 rf ° rf ft S ft ^ P P co "o rffl 1 5 ^ 5 rffl o co cd 5z < H <w cd .rf 1 - 1 O CO -3 u cu H P -.-P P Id ° P o 2 J D C/3 O o O cu CN J rf" 13 ^ P o a p o DH 55 ft-s co O CJ <3 cd >^ Co ^1 C .V rffl S 3 .3 U s ^ ^ o cj S -fl 00 p o ^ . 5 JC' cd P 'Lo c3 rf cd ft 2 o o U £ S cd cd X O & O f -P »-, t>. DH CJ ft "P 2 £ ft-m l < cd - P ft O - I I ^ 9r ^ - P o o DH p -a o CJ . - . <-> - P h >^ DH 2 h J C/3 CU cd -P. cd cd J ^ ^ t , « ft -P XI ^ O D , D . X ft O O DH rf -a 00 O O >>-rf P u u o a T T I,? 1 ,1 I,? I,? ?,l ?,1 1,1 Table 1.2: Reported incidence of clonality in Angiosperm families present in BC. The family name has two superscripts the first being the clonal followed by the aclonal tendency. A superscript 1 designates confirmed presence of each state and 0 confirms the absence of each state while a ? designates that no information was available. For example, Aceraceae has both confirmed clonal and aclonal members while Adoxaceae has clonal members with no information regarding aclonal species. Dicotyledons Aceraceae 1 , 1 Adoxaceae 1 , 7 Amaranthaceae °'' Anacardiaceae 1 7 Apocynaceae '' 7 Aquifoliaceae 1 7 Araliaceae '" Aristolochiaceae 1 0 Asclepiadaceae 1 , 0 Asteraceae ''' Balsaminaceae 0 , 1 Berberidaceae Betulaceae Boraginaceae u Brassicaceae o 1 Buddlejaceae ' Cabombaceae 1 0 Cactaceae '' Callitrichaceae ' ' ? Cannabaceae ' Campanulaceae 1 , 1 Capparidaceae °'' Caprifoliaceae 1 , 1 Caryophyllaceae 1 , 1 Celastraceae 7 , 1 Ceratophyllaceae Chenopodiaceae " Convolvulaceae 1 , ! Cornaceae ''' Crassulaceae 1 , 1 Cucurbitaceae 7 ' ' Cuscutaceae 1 - 1 Diapensiaceae ' Dipsacaeae Droseraceae 1 7 Elaeagnaceae ''' Elatinaceae ''' Empetraceae 1 7 Ericaceae '' Euphorbiaceae 1 7 Fagaceae ''' Fumariaceae Gentianaceae ''' Geraniaceae 1 , 1 Grossulariaceae Haloragaceae '' Hippuridaceae Hydrangeaceae Hydrophyllaceae Hypericaceae 1 , 1 Lamiaceae ''' Leguminosae ''' Lentibulariaceae Limnanthaceae Linaceae ''' Loasaceae 7 ' ' Lythraceae ''' Malvaceae 1 , 1 Menyanthaceae ' Molluginaceae 0 , 1 Moraceae Myricaceae ? ' ' Nyctaginaceae ' Nymphaeaceae '' Oleaceae Onagraceae ''' Orobanchaceae ''' Oxalidaceae ''' Papaveraceae 1 , 1 Parnassiaceae ' Plantaginaceae 1 ^ Plumbaginaceae " Polemoniaceae ''' Polygalaceae ' Polygonaceae ''' Portulacaceae ''' Primulaceae 1 , 1 Pyrolaceae 1 , 0 Ranunculaceae Resedaceae 1 , 1 Rhamnaceae Rosaceae 1 , 1 Rubiaceae 1 , 1 Salicaceae ' Santalaceae 1 0 Sarraceniaceae ' ' ' Saxifragaceae 1 - 1 Scrophulariaceae' Solanacaeae Thymelaeaceae Ulmaceae 7 ? Umbelliferae 1 , 1 Urticaceae ''' Valerianaceae Verbenaceae ' ' ' Violaceae '' Viscaceae ' Zygophyllaceae 0 , 1 Monocotyledons Acoraceae ' Alismataceae 1 7 Araceae 1 , 0 Butomaceae 1 , 0 Cyperaceae 1 , 1 Gramineae Hydrocharitaceae 1 0 Iridaceae 1 0 Juncaceae ''' Juncaginaceae ''' Lemnaceae ''° Liliaceae 1 , 0 Orchidaceae 1 0 Pontederiaceae ''° Potamogetonaceae 1 0 Ruppiaceae 1 0 Scheuchzeriaceae 1 0 Sparganiaceae 1 0 Typhaceae 1 , 0 Zannichelliaceae 1 0 Zosteraceae 1 0 11 Neogene Paleogene Upper Cretaceous Low Cretaceous Jurassic Triassic Perrnian Pennsylvaniart Mississippian Devonian Silurian 0.5 1 Proportion Clonal : Aclonal I Clonal Families El Aclonal Families Figure 1.1: Proportionate representation of clonality among plant families in the fossil record. (Data taken directly from Tiffney and Niklas (1985)). 12 Chapter 2 The Clonal Structure of Maianthemum dilatatum 2.1 Introduction 2.1.1 Determining clonal structure Clonality is an ancestral trait in the plant kingdom (Tiffney and Niklas 1985) and remains a significant component of the life history of many extant Angiosperm species (van Groenendael et al. 1996). Clonal growth has significant demographic, genetic and ecological consequences (Harper 1977) which are still poorly understood. To gain a comprehensive understanding of these consequences, it is necessary to describe the extent of clonality within natural populations. There are various ways to identify genets such as excavation (Cook 1983), compatibility tests in self-incompatible species (Worthen and Stiles 1986), morphology (ICemperman and Barnes 1976) and molecular markers (McLellan et al. 1997). Excavation only measures clonal fragments not genets, so for genet estimation, excavation provides limited information for species with frequent fragmentation. Genet detection from incompatibility tests leads to underestimation when different genets are the same incompatibility type (Worthen and Stiles 1986), or overestimation when incompatibility is variable (Vogler and Stephenson 2001). Morphological traits are generally too plastic to be reliable measures in genet discrimination. Molecular markers are the most efficient and definitive method of genet identification. This study used the multi-locus molecular marker AFLPs (amplified fragment length polymorphisms) which involves the restriction digest of DNA followed by the PCR amplification of select fragments (Vos et al. 1995). AFLPs were selected as they require no prior sequence knowledge and produce many polymorphic loci at a diminishing cost which makes them ideal for genet discrimination (Mueller and Wolfenbarger 1999). 2.1.2 Genotypic and genetic variation within clonal populations Clonal reproduction has often been predicted to result in low genotypic and total genetic variation (McLellan et al.\991). Many predictions are based on the assumption that clonal replication comes at the expense of sexual reproduction and recruitment (Cheplik 1995). Genotypic variation is the number of distinct genets within a population, which is not necessarily equivalent to genetic variation. A population may consist of many genets and as such have a high genotypic diversity but if these genets are highly related then the genetic variation is low (McLellan et al. 1997). 13 Genet dynamics are the cumulative product of local and species-specific characteristics such as site history and sexual recruitment rate. Site history, particularly in terms of disturbance frequency and scale has a significant influence by causing both selective and stochastic genet mortality, thus providing opportunities for genet natality by improving seedling recruitment opportunities. Sexual recruitment is thought to be reduced in clonal as compared to aclonal species as measured by seedling searches (Erikkson 1989) but this generality is based on a limited number of studies. The maintenance of genetic variation within clonal populations is likely due to genet longevity which retards the loss of heterozygosity within these populations (Balloux et al. 2003). Other than genet longevity, there are few hypotheses and even less empirical data regarding how genetic processes such as gene flow, drift and selection may differ in clonal organisms (Eckert 2000). 2.1.3 Spatial arrangement of genetic variation The nature of clonal growth results in an unavoidable spatial structure where clonemates are within some limited distance. Dispersal via clonal growth is typically locally restricted but there are examples of long distance clonal dispersal (Innes 1987, Johansson and Nilsson 1993), found generally in aquatic systems. In local dispersal, the typical distance between clonemates is an ecological balance between close integration and space consolidation versus actively exploring environmental heterogeneity and spreading of mortality risk. These clonal growth strategies have been described by Lovett Doust (1981) in terms of phalanx and guerrilla, referring to plant species with limited to extended distances between ramets respectively. There are no quantitative criteria to designate a species as phalanx or guerrilla but they are useful conceptual descriptions. From the spatial genetic standpoint, the scrambling habit of a guerrilla species would be predicted to result in genetically heterogeneous patches with dispersed genets. Conversely, phalanx species would be expected to have single genet patches with dispersed ramets only found in very old clones. Examples of guerrilla morphologies are Fragaria spp., Linnaea borealis, Trifolium spp. while examples of phalanx morphologies would include Saxifraga spp. and Phlox spp. Spatial structure has often been used to infer population processes in plants. Yet, the mobility of clonal plants complicates any inferences regarding seedling recruitment based on spatial genet structure. If the contribution of clonal growth can be estimated, the remaining 14 genetic structure can be due to limited seed and pollen dispersal distances (Montalvo et al. 1997, Reusch etal. 1999). 2.1.4 Study objectives The size, expanse and relative spatial arrangement of genets are infrequently measured at a fine-enough scale to understand fine-scale ecological interactions (Widen et al. 1994). Yet it has been hypothesized that it is through these fine-scale interactions that a clonal growth habit may differ compared with a unitary species (Handel 1985). Fine-scale measures come at the expense of replication, but can give insight into hypotheses regarding ecological interactions. Therefore, the main objectives of this study were to: 1. Develop AFLP markers for Maianthemum dilatatum suitable for clonal assignment. 2. Determine the fine-scale clonal structure of Maianthemum dilatatum in a single population. Morphologically, M. dilatatum is an example of a guerrilla morphology so it will be determined if this species conforms to the predicted genetic structure. 3. Determine the relatedness of genets within a single population. Comparing the relatedness of genets within the population will yield insight into genet recruitment patterns. 4. Survey the genotypic structure and variability among three distinct populations, allowing the detection of any single genet populations and as a general genetic survey. 2.2 Methods and materials 2.2.1 Study species Maianthemum dilatatum (A. Wood) Nels. & J.F. Macbr. is a common understory rhizomatous herb in lowland and montane areas particularly in riparian areas. It has a distribution ranging from Alaska to Northern California and eastward to Idaho (Stewart 1994, Douglas et al. 1997). Maianthemum dilatatum can be a dominant understory species in moist areas within Sitka spruce, Alder and Douglas fir forests, forming extensive continuous cover (Kawano et al. 1971). Maianthemum is a rhizomatous species, with a lateral bud associated with every root whorl. Buds do not develop sequentially along a rhizome but remain dormant until activated whereupon they initiate a rhizome branch extension or turn upwards becoming a ramet. In this thesis, a ramet is defined as the leaf or floral stalks. Non-reproductive ramets consist of a single cordate leaf, leaves die back in late fall but ramets are perennial with infra-petiolar buds holding the next season's stem. Young ramets tend to be non-reproductive initially but seem to become capable of producing flowering stems after approximately their third to fourth season. Flowering ramets consist of three, rarely four alternate leaves with a raceme of small, white flowers. Floral stalks have no infra-petiolar bud 15 and the subsequent years' stalk arises from a lateral bud. Maianthemum dilatatum is an interesting study species given the large continuous patches of which the clonal structure is unknown. Additionally, to date no molecular work has ever been done on this species. 2.2.2 Study site and sampling The primary study site was in Pacific Spirit Park located on the westside of the city of Vancouver, British Columbia. Within the primary site, this study was constrained to a total area of three ha, where twenty-one patches of M. dilatatum were found, collectively covering 3236m2. Patches were defined as: continuous M. dilatatum cover with no breaks greater than 2-m. This study plot was considered to be a sub-population, as no other major patches occurred within 200-m to the west, north or south. Approximately 30-m east from the last study patch, there was a large patch with less than 25 floral ramets occurring, but beyond this all patches were small until beyond 200m to the east. Patch margins and sizes were determined using a tape measure and compass in relation to a lattice of reference points 1 established throughout the study area. In the Pacific Spirit subpopulation, 8-m transects which were 8-m apart were established in every large patch. Patches too small for this scale had samples taken from all edges that were greater than 2-m apart. The choice of grid size was a balance between budget and biology, as M. dilatatum clonal fragments can average up to 1.8-m in length (Lezberg et al. 1999). At each sampling point, the leaf was taken from the nearest ramet. Systematic sampling was chosen as it is preferable for autocorrelation analyses (Epperson 1993), provides a two-dimensional measure of genet size and allows for inference about ecologically relevant intermingling (Widen el al. 1994). Fifteen pairs and three triplets of known clonemates were collected, by excavating small areas until ramets could be found which were clearly connected. 2.2.3 Estimates of sexual recruitment A first-year M. dilatatum genet can be recognized by the absence of any leaf scars, no other connected ramets and a leaf width of 5 mm to 10 mm. The leaf size was determined by observation of individuals raised from seed and therefore known to be first year seedlings. Using leaf size as an indicator, one year old recruits were searched for along all of the established 8-m transects. These yearlings were not disturbed aside from the collection of a small leaf sample for genetic analysis. Detection of first year recruits is underestimated, given their small size and the depth of leaf litter. Under the prior observation that seedlings are rare, adopting a large-scale sweep would increase the probability of detection as opposed to a quadrat search scheme. 16 2.2.4 Population level comparisons of genotypic and genetic variation Several populations were chosen for estimates of genotypic and genetic variation. These populations were located at Burnaby Lake (BL) and two sites in the Lower Seymour Conservation Reserve which were separated by approximately six km (LSCR and SY). In these populations, samples were taken along a 30m transect such that most large patches would be intersected. This scheme resulted in 11 samples from BL, 17 samples from LSCR and 11 samples from SY. 2.2.5 DNA extraction for leaf tissue. Leaf samples were collected, washed and stored at -80°C until DNA extraction. Total genomic DNA was extracted using a modified CTAB (Cetyltrimethylammonium bromide) extraction protocol (Porebski el al. 1997). 0.25g of leaf tissue was ground to a powder using liquid nitrogen and a chilled mortar and pestle. Grindate was then added to 20mL of extraction buffer (2% CTAB, 1.4M NaCl, 0.1M Tris-HCl, 20mM EDTA (Ethylenediaminetetraacetic acid), 2% P-mercaptoethanol and 1% PVP (polyvinylpyrrolidone MR 40, 000)) and incubated at 65 °C for one hour. Two chloroform isoamyl alcohol (24:1) equal volume extractions followed using a 20 minute spin at 3000rpm. DNA in the supernatant was precipitated overnight at -20°C using a 2/3 volume of isopropanol. DNA was pelleted in a 30 minute spin at 4, 000 rpm at 4 "C. After discarding the isopropanol, the resulting DNA pellet was washed with 70% ethanol. DNA was redissolved in 500ul of sterile, distilled water. DNA was then treated with RNAse (1 Oug/mL) for lhour. Proteinase K treatment (1 Oug/mL) followed for 45 minutes. Two equal volume extractions of 1:1 phenol and chloroform:isoamyl followed, with a final 1:1 chloroform:isoamyl extraction. DNA was precipitated with 0.15M NaCl and 2 volumes of 100% ethanol. DNA was spun down at 10, 000 rpm for 25 minutes followed by two washes with 70% ethanol. DNA was resuspended in deionized water and was left overnight at 4°C before use in the AFLP protocol. Template quality and quantity were measured using a spectrophotometer. Only samples which were of sufficient quality as defined by an absorbance ratio (A26o/28o) of 1.6-1.9, were used in further analyses. Samples were all diluted with filter-sterilized, deionized water to a standard concentration of 50ng/ul. 2.2.6 AFLP protocol This study suffered a lack of AFLP reproducibility, which was resolved by a re-collection and re-isolation in 2003 with a protocol targeting recalcitrant species. The source of error in this work was confirmed to be polysaccharide and polyphenol contamination, since reproducibility 17 was only obtained after the use of a rigorous DNA extraction protocol targeting both of these contaminants. The AFLP protocol was performed following Vos el al. (1995) with the exception of using near infrared fluorescence to label the EcoRI final amplification primers. 500ng of template DNA was digested for 2.5 hours with 5U Msel, 7.5U EcoRI and 4ul restriction -ligation (R/L) buffer (50mM Tris-Hac pH 7.5, 50mM MgAc, 250mM K Ac, 25mM DTT) to a final volume of 18.75ul. Subsequent ligation reactions proceeded for 5 hours at 37°C with 0.3 Weiss U of T4 DNA Ligase with 1 OmM ATP, 15.6 pMol Mse adapter and 1.56 pMol EcoRI adapter, 0.63ul of R/L buffer made to a final volume of 15.6ul. Preamplification was carried out using +2 pre-selective primers. Reaction mixtures were 12.5 ul with 3.Out 1:10 diluted ligated fragments, 0.2mM dNTPs, 5uM of a +2 EcoRI and Mse primer, Roche PCR Buffer (lOOmM Tris-HCl pH 8.3, 500mM KC1, 15mM MgCl2) and 1.2U of Taq polymerase. PCR reactions were carried out on MJ thermocyclers for 28 cycles of 30s denaturation at 94°C, 30s annealing at 60°C and a 60s extension at 72°C. Four primer combinations were used for the selective amplification, with three and four selective nucleotides for EcoRI and Mse repectively (E+3/M+4). Final amplification reaction mixtures were as follows: 3.0ul 1:10 diluted ligated fragments, 0.2mM dNTPs, luM of a +3 700 or 800 IRD labelled EcoRI primer (Li-COR), 8.3 uM of a +4 Msel primer, 1 ul Roche PCR Buffer (lOOmM Tris-HCl pH 8.3, 500mM KC1, 15mM MgCl 2) and 0.6U of Taq polymerase). The PCR reactions were carried out on a MJ thermocycler for 38 cycles of 30s denaturation at 94°C, 30s annealing and a 60s extension at 72°C. The initial cycle had an annealing temperature of 65°C which decreased 0.7°C/cycle for 12 cycles followed by an annealing temperature of 57°C for 23 cycles. Primer Screening and Selection: Forty-five primer sets for both Pst/Mse and EcoRI/Mse enzyme combinations were screened for polymorphisms, evenness of fragment distribution and reproducibility. The final primer combinations selected were M+CCAG/E+AGC, M+CCCA/E+AGG, M+CCGG/E+AGG and M+ATGT/E+CCC. Pst was avoided due to its methylation sensitivity. The four combinations of AFLP primers generated 243 unambiguous, polymorphic markers with an average polymorphism of 76.5 %. 2.2. 7 Detection and scoring of AFLP Fluorescently labelled fragments were fractionated on 25cm 6% Long Ranger polyacrylamide, 7.5M urea and IX TBE (89mM TRIS, 89mM boric acid, 2mM EDTA). 6ul of loading buffer was added per sample (95% filter-deionized formamide, 20mM EDTA pH 8.0 and 18 lmg/mL bromophenol blue (USB)) followed by a three minute denaturation at 95°C. 4ul of product was loaded using an eight channel Hamilton multiple pipette. An IRD-labelled molecular weight standard (50-700 bp-LiCor) was loaded in the first and last lanes. Gels were run using IX TBE buffer on LiCor 4200 automated sequencers with operating settings of 2000V, 35mA, 70W, 50°C, motor speed 4 and 16-bit pixel resolution. Gel image TIFF files were scored manually but SAGA AFLP MX software (LiCor) was initially used for band sizing. Bands between 50 and 700 basepairs were scored as being present (1), absent (0) or in a few cases undetermined (9) with no consideration given to intensity. In this study, only loci which were 1) unambiguously present or absent across all individuals and 2) in frequencies <98% as per Lynch and Milligan (1994) were used in further analyses. Pruning has been demonstrated by Isabel et al. (1999) to reduce the bias associated with dominant markers in estimating genetic parameters such as Nei's gene diversity and FST-2.3. Data analysis 2.3.1 Genotypic diversity Pair-wise differences in AFLP profiles between all sampled ramets were determined using Jaccard's similarity index, calculated as Sj=a/(a+b+c) where the letters refer to the band states of individuals being compared such that a=l/l, b=0/l, c=l/0. The Jaccard coefficient is preferential for AFLP studies because negative matches (d=0/0) are not considered to be indicative of similarity (Sneath and Sokal 1973, Brown 1996). AFLP banding patterns are rarely identical even among replicates of a single individual (Arens et al. 1998; Winfield et al. 1998) such that criteria need to be established to allow for variability within a genet. To have an objective method of genet assignment, a similarity threshold was established. The first was based simply on the variance between known clonemates collected as described in 2.2.2. The second was a threshold proposed by Douhovnikoff and Dodd (2003), which incorporates sibling data. The Douhovnikoff and Dodd (2003) threshold (T) is calculated as T=(ocfrs+as[ic)/(cs+ac) where u,s,p,cand os, oc are the means and standard deviations of known sibling and clonemate pair-wise similarity distributions. A threshold based on intra-genet variability is superior to any criteria based on intra-ramet variability as it incorporates the variability due to isolation artefacts, somatic mutation and microbial contamination. The known similarity distributions were based on twenty-one pairs of physically connected clonemates collected as described in 2.2.2 and ten sibling arrays obtained from open pollinated, unbagged plants from 3.2.2. 19 The discriminatory ability of the marker influences the power to delineate genets (Avise 1994), so the power of the AFLP markers, was estimated as the probability that a particular AFLP profile would re-occur by chance. The chance probability of an AFLP profile was calculated for each genet according to Sydes and Peakall (1998) for dominant markers such that Pdgen=FIpi where pj is the band frequency at the genet level. The lower the frequency of a marker, the greater the likelihood that individuals sharing this band state are related. Genetic fidelity within a genet may be masked by protocol artefacts or disrupted by somatic mutation such that clonemates may appear to be different genets. Differences due to recombination can be differentiated from mutation events using character compatibility regardless of marker dominance (Mes 1998). In a dominant marker the possible states between two loci are 11,01, 10 and 00, where the presence of all states between four individuals is improbable without recombination (Eckert et al. 2003). Therefore, recombination events will manifest as matrix incompatibility counts (MIC) between individuals (Mes 1998, Wilkinson 2001) where pure clonality would yield a MIC=0 (Eckert et al. 2003). The contribution of particular genotypes to total MIC can be estimated by successively removing the most incompatible genotypes until the MIC=0 where the remaining genets are possible clonemates and misidentified by the Jaccard distance criteria (Eckert et al. 2003). Matrix incompatibility counts and genotype contributions were calculated using the program PICA 4.0, (Wilkinson 2001) based on 50 genets and 100 loci. Measures of genotypic diversity are only informative when compared among equivalently sampled populations (Barbour et al. 1987) and as such will be reported for the Pacific Spirit population only. The Simpson's Diversity Index (D) was calculated with corrections for finite population size (Pielou 1969). Simpson's diversity index is a probabilistic measure of a repeated sampling of a particular genet within a population, which is calculated as D=l -{[X«i("i -1)]/[N(N-1)]} where N is the total number of sampled ramets and n-, is the number of ramets with a given AFLP pattern /. Therefore, a monoclonal population would have a D=0, while a population where all ramets are from distinct genets would have a D=l (Pielou 1969). Additionally, the Genotypic Evenness Index (E) was calculated (Fager 1972). The Genotypic Evenness Index is a measure of the disproportionate size/representation of the sampled genets where E=0 would describe a population with complete dominance of a single genet, and E=l would indicate that genets have similar numbers of ramets. Genotypic evenness is calculated as: E= (D o b s-Dm i n)/(Dm a x-Dm i n) where Dfnin=[(G-l)(2N-G)]/[N(N-l)] and Dmax=[N(G-l)]/[G(N-l)] and G=the number of genets detected and N is the number of sampled ramets. 20 2.3.2 Genetic variation within the Pacific Spirit subpopulation As suggested by Lynch and Milligan (1994), to reduce the bias associated with low frequency null alleles, only loci which had frequencies of less than 1 -(3/N)% were incorporated into genetic variation measures, where N is the number of ramets sampled. Gene diversity within populations was estimated using average expected heterozygosity as per Lynch and Milligan (1994) calculated with AFLP-Surv version 1.1 (Vekemans 2002) with the Bayesian estimator with non-uniform allele frequency. The Shannon Index of phenotypic diversity (Hs) was also calculated where: H s =-XPilog2Pi (Lewontin 1972). Relationships among genets were estimated using Hardy's (2003) relatedness estimator for dominant markers r^. These relatedness measures were calculated using SPAGeDi version 1.1 (Hardy and Vekemans 2002). 2.3.3 Spatial structure of genetic variation within populations Spatial autocorrelation analysis was done within all populations to detect any correlation between genetic and geographical distance between the ramets. Sampled ramets were mapped so these coordinates could be used to generate a physical distance matrix. The physical distance matrix and the Jaccard distance matrix were then subject to a Mantel test using the program GenAIEX v.5.1 (Peakall and Smouse 2001). The Mantel test is a matrix correlation which describes the degree of association between two distance matrices, described as a product moment correlation r, the significance of which is tested through permutations (Sokal and Rohlf 1995). The Mantel test incorporated samples taken beyond the subpopulation. A multivariate version of the Mantel test was also used to estimate how the correlation changes over distance classes within the Pacific Spirit subpopulation, as developed by Smouse and Peakall (1999). The distance at which the plotted correlation crosses the x- axis corresponds to size of the 'genetic patch'. The genetic patch is the distance where localized seed and pollen dispersal occur (Chung and Epperson 2000). Distance classes were set by equalizing sample number within a distance interval to incorporate a sufficient number of comparisons (Epperson and Li 1997), which reduces standard error. The multivariate Mantel test was done at two levels in an attempt to separate the contribution of clonal growth and gene flow patterns to the genetic structure within the subpopulation. When all ramets, including clonemates are considered individually, the resultant structure is the product of both clonality and gene flow. The influence of clonality can be isolated if an average coordinate position is given to those genets which had been detected 21 several times (Reusch et al. 1999). The averaged coordinate was calculated for a genet with n detected clonemates: (x,y)genet= (Exi+—+xn)/n, (Eyi+—+yn)/«. Further analyses involved an AMOVA (analysis of molecular variance) (Excoffier et al. 1992), to determine if variability in genetic distance between genets was greater within or between multi-genet patches. When patches are suspected to have some biological relevance, AMOVAs have been done at the patch level (Ingvarsson and Giles 1999, Torimaru et al. 2003). AMOVA requires Euclidean distances so AFLP-based distances were calculated according to Huff et al. (1993). The AMOVA analyses were carried out using the GenAIEX program v5.0 (Peakalland Smouse 2001). 2.3.4 Population level comparisons of genotypic and genetic variation All population samples were subject to most of the same analyses as detailed above: the Jaccard similarity threshold for clonal assignment, r ab calculations, and the Mantel multivariate correlation. 2.4 Results 2.4.1 Genotypic diversity The similarity threshold as per Douhovnikoff and Dodd (2003) was calculated to be T=0.945. The frequency distributions of similarities for clonemates and siblings are given in Figure 2.1. Using this criterion, within the Pacific Spirit site 74 putative genets were found, eight of which had multiple clonemates detected 2 to 3 times at the 8-m sampling scale. Patch characteristics are listed in Table 2.2. The highest probability of randomly drawing a particular single AFLP profile was P d g c i ^ l -69E"31. This measure assumes random mating and independence among all loci. The Jaccard distance values of the genets are given in Appendix A. Character compatibility analysis results showed a high MIC of 10,349 within the population. Clonemates were tested and showed a MIC =0. Putative genets were tested for character compatibility but all genets needed to be deleted before the MIC count was reduced to baseline counts (MIC=0) (Fig. 2.2). The MIC is not directly proportionate to sexual reproduction (Mes 1998) but because genets are necessarily separated by recombination events, comparison among them should result in high MIC. Simpson's Diversity Index (D) and Genotypic Evenness Index (E) were estimated to be 0.99 and 0.96 respectively. Five, one-year old recruits were found throughout the entire study population which is an acknowledged underestimate. Their recruitment locations are mapped as 22 enclosed crosses in Figure 2.4. The DNA isolated from these individuals resulted in failed AFLP reactions due to damaged leaves. 2.4.2 Genetic variation within the Pacific Spirit subpopulation Nei's gene diversity within Pacific Spirit was calculated to be Hj=0.30 with a standard error of 0.102. Shannon's phenotypic diversity index was calculated to be 0.322. According to theoretical expectations, rab >0.50 is indicative of parent-progeny or full-sib relationships while rab >0.25 are indicative of half-sibships (Lynch and Milligan 1994). The distribution of r^ values is given in Figure 2.3. The distribution was normal as calculated by the Lilliefors' test for normality (pO.OOl). 2.4.3 Spatial structure of genetic variation within populations Genet extent was variable, with the average distance being 11.76-m, and the greatest being 30.17-m. By our sampling scheme genets appeared to be physically close within patches and patch margins were only representative of genet margins in the small patches. Patch sizes 2 2 ranged from lm to 1043 m (as listed in Table 2.2) with the larger patches appearing as an amalgamation of multiple genets. The Spearman's coefficient of rank correlation, between patch area and genets detected was rs=0.870 pO.OOl. The Mantel correlation between Jaccard distance and geographical distance matrices was r=0.1249 p<0.05 as shown in Figure 2.5. The multivariate Mantel test allowing for clonemate joins was higher at the smallest distance classes (rO.289 pO.OOl), as compared to structure where clonemate joins and thus clonal influence was removed (r=0.209 pO.OOl). The genetic patch width was approximately 55-m, whereupon any statistically significant correlation between genetic and geographical distance is lost. AMOVA analyses were carried out at the genet level on all multi-genet patches (Table 2.3). The results demonstrated that multi genet patches are not significantly differentiated from other genets in other patches since the majority of the molecular variance (77%) could be found within patches. 2.4.4 Population level comparisons of genotypic and genetic variation In all three populations, all transect points picked up different genets. Nei's gene diversities were calculated for BL, LSCR and SY as 0.236, 0.218 and 0.249 respectively. Given in the same order, the Shannon diversity indices were: 0.242, 0.233 and 0.259. The average rab within BL, LSCR and SY were 0.205 SDO. 117, 0.208 SDO. 186 and 0.171 SDO.093 respectively. Under an ANOVA analysis, these relatedness measures were not significantly 23 different p>0.05. Significant spatial Mantel correlations were only found in LSCR <>2=0.05 p=0.02). 2.5 Discussion 2.5.1 AFLP reproducibility AFLPs are considered to be very reproducible markers, as demonstrated in Powell et al. (1996), where AFLP band patterns are largely consistent among replicates and robust to reaction conditions. The lack of reproducibility among replicate AFLP profiles was a significant problem initially in this study such that it is worth briefly discussing. Initial reproducibility among replicates from a single individual from the same extraction yielded an unacceptable range of approximately 75-80% similarity. For AFLPs, the limiting factor in achieving reproducible profiles is DNA template quality. Any degradation of DNA through freeze-thaw cycles of tissue, action of endogenous nucleases or mechanical shearing will reduce AFLP reproducibility. Secondary metabolites also affect template quality, which can be a significant hurdle in species with no prior molecular work, since it would be unknown to the researcher what inhibitory compounds may be present. The propensity of plants to produce secondary compounds makes them notorious for difficult DNA isolation (Baker et al. 1990) but related difficulties have also been reported in invertebrates and fungi (Rudi et al. 1997). Polysaccharides and polyphenolics are commonly reported as inhibitors of molecular procedures by complexing with the DNA or interfering with enzymatic processes (Do and Adams 1991, Richards et al. 1994, Koonjul et al. 1999). Any interference with the restriction enzyme digestion creates a heterogeneous mix of fragment sizes. The variation among these fragments is artifactual and will manifest as a lack of reproducibility between replicates of the same extraction. In this study, measures of DNA quality such as spectrophotometric ratios of A260/280, digestability tests, were not indicative of the presence of inhibitory compounds. In fact, certain polysaccharides have similar structure to nucleic acids and will register as high DNA concentrations with a moderate A260/28O ratio. Neither post-isolation genomic cleaning using a modified CTAB protocol nor diluting the template as per Rudi et al. (1997) were successful. Additionally, attempts with commercially available DNA isolation kits such as Qiagen DNeasy were unsuccessful because buffer concentrations cannot be modified. Although scaling down tissue within these kits is possible, in the case of Maianthemum, tissue volumes became too small (<50mg) to yield sufficient DNA. 24 After exhausting both AFLP procedural modifications and D N A clean-up protocols, the resolution came in the form of a complete recollection in 2003 and the adoption of a protocol targeting recalcitrant species. The key elements were the implementation of 1% PVP (polyvinylpyrrolidone) and high salt, which are essential to deal with polyphenolics and polysaccharides respectively. A second key element was adopting a very high ratio of buffer to tissue (25mL: 0.25g) and using a total of five organic extractions. Recalcitrance can be a seasonal phenomenon for some species or a problem with certain tissue types, but in general young tissues are recommended to avoid problems (Jobes et al. 1995). In this study, even young tissues required the same effort in D N A isolation in order to achieve reproducibility. A second area that may lead to higher levels of variability within samples is a large genome size or high polyploidy. No genome size estimates exist for Maianthemum, but relatives Smilacina spp. and Convallaria spp. range from 10 878 Mbp to 24 157 Mbp (Bharathan et al. 1994, Bennett and Leitch 2003) which in the plant kingdom are very large. Large genomes are problematic in PCR reactions simply due to the excessive number of fragments which encourages mispriming and lowers the amplification efficiency (Garner 2002) which could manifest as a lack of reproducibility in AFLP profiles. As AFLP gains popularity for species with no previous molecular work, an increasing number of reports of AFLP difficulty are expected to arise particularly with plants. Several precautions are recommended: 1) Initially implementing the most rigorous isolation protocol, then scaling down accordingly; the AFLP protocol is robust to DNA quantity but not quality. 2) Reproducibility tests should be nested in primer screening so that problems can be detected early. If possible, reproducibility should be confirmed prior to extracting DNA from the entire collection, regardless of the perceived quality because post-isolation cleaning attempts are rarely successful (Arens pers. comm.) such that all isolated samples may be unusable. 2.5.2 Genotypic diversity and relatedness among genets Seventy-four distinct genets were found within the Pacific Spirit subpopulation (Fig. 2.4). In terms of clonal assignment, there were five ambiguous cases where similarities were outside the threshold but were as close as 92%, which is tempting to suggest that they represent a single genet. Using a threshold allows an objective determination of intra-clonal variability and/or separating closely related individuals and clonemates through comparing similarity distributions. The drawback of the threshold that it's accuracy is dependent on the distribution of the particular sibling array, which can be a problem when using a sibling array from open pollinated plants. 25 The calculated threshold value of 0.945 corresponded closely to the greatest variability demonstrated within a genet (0.940). The threshold based on clonemate variability served as our threshold because sibling data were thought to be unreliable (see 3.4.4). There are several reasons for not relaxing the criterion: 1) the majority of pair-wise similarities between clonemates were <5% different 2) if the threshold was relaxed, the 'fringe' clonemates would not have consistent clonal similarities among all other putative clonemates in a genet group. Additionally, Douhovnikoff and Dodd (2003) reported Jaccard coefficients as high as 0.98 between siblings, which makes any threshold below 0.94 difficult to justify. The sibling distribution was based on unknown paternity but if controlled crosses were in place then the threshold could be better structured. The observed values of D and E of 0.99 and 0.97 suggest a diverse and symmetrically sized population. These diversity values are high, compared to the average values among plant species complied by Widen et al. (1994) which was a D=0.75 (range of 0.13-1.0). Although measures of D and E are often included in clonal structure studies, similar sampling schemes are necessary for valid comparison (Barbour et al. 1987, McLellan et al. 1997) such that perhaps standardizing for sampling intensity could remedy this problem. The number of genets detected is a product of both the discriminatory ability of the marker and the sampling scheme used. For clonal assignment, polymorphic loci are needed to discriminate between close relatives and clonemates, which can be obtained through a few low frequency markers or many high frequency markers. Studies have shown that increasing the number of markers can lead to genet reassignment, generally as distinct genets. Obviously, this plateaus and the only way to find additional genets is to increase the sampling effort. In this study the number of loci was reduced from 447 to 243, which had no effect on the clonal assignment decisions, so the number of loci are believed to be sufficient. Loci were reduced so as to include only the strong and unambiguous bands. Within our population there was 3236m of M. dilatatum cover with a 100% cover density, there could be up to 70-100 ramets/ m 2 so there are thousands of ramets, of which a fraction was sampled. If additional ramet sampling had been done in this population, undoubtedly more genets would be found. In fact, all genets detected in Table 3.1 were distinct from any genets shown in Figure 2.4 within the patch where the plot was taken. Since Janzen (1977) and Loveless and Hamrick (1984) suggested that clonal populations should demonstrate lower genetic variability, many studies have demonstrated this pattern is not universal or even common. Yet, some clonal populations do have very low genotypic variations 26 such as seven genets for a population of Haloragodendron lucasii (Sydes and Peakall 1998) or even one in Decodon verticillatus (Eckert and Barrett 1993). Population-level processes specific to clonal populations are not well understood, since both ramet-level and genet-level dynamics need to be concurrently considered. Genet-level dynamics are difficult to measure due to the necessity of using molecular markers and the long time scale under which changes occur. These limitations are increasingly overcome by the availability of more financially feasible markers and the implementation of population simulations (Eriksson 1994). Genotypic diversity within a clonal population, as in any other population, is the product of genet natality and mortality but clonal populations differ in these demographic parameters. Sexual recruitment in clonal plants is generally assumed to be low (Erikkson 1989, Schmid 1990), such that diversity is hypothesized to be preserved by high genet longevity (Erikkson 1993). Simulations assuming long genet longevity by Watkinson and Powell (1993) support this suggestion that very low seedling recruitment rates could be sufficient to maintain genetic variation within clonal populations. Yet, seedling recruitment does occur to some degree since studies of clonal structure generally find multiple genets in natural populations. Inferring genet dynamics from genotypic diversity needs to be considered over a longer time scale which has been best explained by Erikkson (1993) and Erikkson and Froborg (1996) in terms of a gradient of recruitment patterns of initial seedling recruitment (ISR), repeated seedling recruitment (RSR) and recruitment in windows of opportunity (RWO). In the ISR a decline would be expected in genotypic diversity over time, due to morality factors and competitive exclusion which was reported in Lolium perenne (McNeilly and Roose 1984) and Circaea luteliana (Verburg et al. 2000). Conversely, the greater influx of genets with RSR and RWO strategies would result in stable or increasing genotypic diversity. The RSR strategy is seemingly common, as it has been reported in 40% of 68 forest species (Erikkson 1989) and in other herbs such as Viola riviniana (Auge et al. 2001) and Anenome nemorosa (Holderegger et al. 1998). In fact, this repeated seedling recruitment balances genet mortality within natural plant populations (Bierzychudek 1982), and contribute equally to population growth as compared to clonal recruitment (Inghe and Tamm 1985). In the Pacific Spirit subpopulation of M. dilatatum, sexual recruitment appears to occur repeatedly such that the RSR or RWO strategy could be operating. Indirect evidence for this is the detection of multiple genets of varying expanse, the observation of germinating seed in December 2002 and the detection of five, one year-old genets in the 2003 field season. Conversely, several studies of clonal plants have 27 reported limited to complete absence of genet recruitment within the study time frame (Ellstrand and Roose 1987, Hamrick and Godt 1989, Widen et al. 1994). The presence of one-year recruits is of importance in that it demonstrates the potential for sexual recruitment, but the genets need to survive for effective recruitment to take place. Seedling recruitment patterns are affected by both seed availability and but also by opportunity, such as the existence of safe sites. Kudoh et al. (1999) report in Uvalaria perfoliata, that genotypic diversity was much higher in canopy gaps with only single genet patches being found in closed canopy sites. It is feasible that a similar process could operate for Maianthemum dilatatum in light-limiting sites, as fruit does not appear to set in the dense shade (van der Kamp pers. comm.). Recruitment in Maianthemum dilatatum seems to be favoured in moist sites and in dead wood wherein three of the five yearlings were found. Seedling regeneration dynamics in clonal plants should only differ from aclonal species if ramets competitively exclude seedlings (Abrahamson 1980, Erikkson 1997). Yet, there have been suggestions that there may be recruitment niches for seed and ramets such that each can recruit in microsites where the other cannot (Abrahamson 1980). Initial ramet recruitment success is greatly facilitated by physiological integration and the avoidance of juvenility so that harsher microsites may be inhabited. However, the initial survival probabilities of a genet are equivalent to that of an aclonal genet up until the production of the first clonemate. Genet longevity is the cumulative probability of individual ramet survivorship, such that increasing the number of ramets constituting a genet, reduces the mortality risk (assuming independence of ramet mortality) (Harper and White 1974, Tanner 2001). The effectiveness of mortality risk spreading is a function of the scale of the mortality factor such that clonal dispersal would only be effective against patchy and local factors (Sackville Hamilton et al. 1987). A patchy mortality factors for M. dilatatum in the Pacific Spirit population would be falling trees, which are frequently found within patches, with at least two new events in 2003. Another significant source of ramet mortality at this site is off-trail trampling. Attempts to age genets are based on relating genet area to growth rates which has been used for genet longevity estimates in woody or slow-growing species. Extensive genets of Populus tremuloides were estimated to be 1 million years old (Kemperman and Barnes 1976, Mitton and Grant 1996) while Carex spp. were aged to 2000 years (Steinger et al. 1996). Generally, herbaceous genets cannot be aged reliably since the growth patterns are less conserved and transient so only minimal ages can be estimated. Yet, comparatively minute genet longevities of 3-10 years have been estimated for Trientalis europaea and Asarum 28 canadense (Piqueras and Klimes 1998, Damman and Cain 1998) and in Circaea lutetiana (Verburg et al. 2000). Genet longevity in M. dilatatum cannot be estimated, but two separate ramets were found in 2002 which were estimated to be greater than 15 years old on the basis of leaf scars. In 2002, markers were placed at the margins of some patches but in 2003 no leaves occurred past these markers so a yearly growth rate was not obtained. Given their enhanced survival, clonal species may seem to be less vulnerable to extinction factors. But simulations predict that the effects of reduced recruitment imposed for example by fragmentation may not manifest in population dynamics for 50-100 years (Erikkson 1994). Erikkson (1994) also suggests that this scenario would result in a failure to detect declines in population viability until extinction was inevitable. Ironically, species which form large genets would be at the greatest risk to be incorrectly assessed as being common and would require a larger minimum habitat area to accommodate the sufficient number of genets to maintain genetic viability. 2.5.3 Genetic variation within the Pacific Spirit subpopulation Both Nei's gene diversity and Shannon's phenotypic diversity index were high as compared to summarized reports of diversity (Hamrick and Godt 1996, Gaudeul et al. 2000). High phenotypic diversity can be indicative of low levels of inbreeding but is also a predicted pattern from a clonal habit (Balloux et al. 2003). Determining the relatedness of genets allows for some inference regarding founder events and gene flow. Founder effects can significantly affect the genetic trajectory of a population especially in the absence of gene flow. In situations with repeated recruitment coupled with gene flow, these founder patterns become obscured and hard to detect with typical sampling intensities. The range of relationships among genets and within patches (Fig. 2.3, Table 2.2) suggest that both local recruitment and mating occurs but also that non-relatives are present given the negative rab values. Hardy (2003) determined through simulation, that the rab between known relatives often deviates from the theoretical expectation of rab- Theoretical expectations are derived from Mendelian transmission probabilities such that the rab for first-order relationships of parent-offspring and siblings is expected to range from 0.5 to 0.25 respectively with non-relatives having r a b values of <0 (Lynch and Walsh 1998). Lynch and Milligan (1994) also determined that r a b distributions for different degrees of relatedness may overlap such that they cannot be differentiated. In this study, there was significant overlap between the distributions of pair-wise relatedness within the Pacific Spirit genets to that of non-relatives. The non-relative data came 29 from pair-wise comparisons between genets from different populations, which are almost certainly unrelated. Therefore, although it can be determined that there is a range of relatedness within the Pacific Spirit subpopulation, genets cannot reliably be classified into genealogical relationships. Despite this uncertainty, the relatedness values were calculated in addition to the Jaccard coefficient because they take into account allelic frequency. A different approach to infer recruitment has been to use maternally inherited genomes such as those of the chloroplasts and mitochondria from which maternal lineages can be determined. This is a powerful tool to separate seed and pollen mediated gene flow, which differ in dispersal dynamics but can be difficult to separate in natural populations based only on nuclear D N A (Ziegenhagen et al. 2003). 2.5.4 Spatial structure of genetic variation within populations Spatial structure within clonal populations is formed through clonal growth, seed recruitment patterns and spatial heterogeneity in selection (Ennos 2001). Long distance clonal dispersal is also possible through vegetative propagules or through apomictic seed (Avise 1994) with reported distances of 1.5 km in Ranunculus lingua via waterways (Johansson and Nilsson 1993) to 150 km in Enteromorpha lima (Innes 1987). Generally, clonal dispersal is restricted such that the spatial distribution of clonemates is the result of clonal morphology and its responsiveness to conditions in the microenvironment. In a guerrilla species, ramet dispersal can be extensive particularly across poor habitat. With respect to genetic spatial structure, the increased dispersal spreads the high genetic similarity across a greater geographical distance, resulting in less of a contribution to fine-scale spatial correlations. Species in the phalanx range would have ramets much more clustered such that a steep structure would be found, very high but dropping beyond genet margins. Maianthemum dilatatum is of a guerrilla morphology and as predicted clonemates were spatially dispersed with detected clonemates being as far as 30.17-m apart. Clonemates being separated by moderate distances results from clonal dispersal, which has similarly been reported in other forest species such as Vaccinium stamineum (Kreher et al. 2000) and Anenome nemorosa (Stehlik and Holderegger 2000). The dispersed genets and relatively long-lived ramets found in this study are similar to the structure and ramet longevity found in forest plant species in a community wide analysis by Tamm et al. (2001). Genets intermingle do at the centimetre scale which would not have been apparent from the 8-m sampling scale. Despite intermingling within these 1-m plots, commonly only single genets or at most three genets were detected (Table 3.1) which is in contrast to other studies of 30 clonal herbs that found higher diversity at this scale (Kudoh et al. 1999, Stehlik and Holderegger 2000, Auge et al. 2001, Ziegenhagen et al. 2003). Generally, gene flow declines with distance, which establishes a correlation between geographical distance and genetic distance resulting in spatial genetic structure. Most studies find weak spatial structure and/or that this structure only exists over small spatial scales (Heywood 1991, Smouse and Peakall 1999). The detection of spatial genetic structure depends on the spatial and temporal scale and patterning of the genetic variation (Smouse and Peakall 1999). Depending on whether clonemates are clustered or dispersed, clonality should contribute either substantially at a fine scale or minimally over a larger scale (Wells and Young 2002). Clonality was the primary factor behind spatial structure in Zostera marina (Reusch et al. 1999) and in Quercus chrysolepis (Montalvo et al. 1997), but was inconsequential in Eurya emarginata (Chung and Epperson 2000). Within M. dilatatum, clonality does contribute at the shorter distance of approximately 25-m which corresponds closely to the largest distance between clonemates (Fig. 2.4). The structure found in M. dilatatum which was not attributable to clonality, could be generated by pollen and seed-mediated gene flow. Seed dispersal distances in the sympatric herbs Cryptotaenia canadensis, Osmorhiza claytonii and Sanicula odorata coincided closely with observed genetic structure (Williams 1994). Restricted seed dispersal is expected to contribute to genetic structure to a greater extent than pollen because seeds contain double the genetic contribution (Doligez et al. 1998). Restricted seed dispersal would also manifest as strong spatial structure in maternally inherited genes (Heuertz et al. 2003). The genetic patch size of 55-m is larger than other herbs and is more in the range reported for trees. But at present, the genetic patch widths cannot be compared because: 1) the level and extent of the underlying spatial structure may be obscured by statistically weak tests or sampling scale (Smouse and Peakall 1999) and 2) no other studies have used AFLPs and the multivariate Mantel test to detect genetic structure. Ziegenhagen et al. (2003) used AFLPs and the change in average Jaccard distance over geographical distance in Galium odoratum, within which they found low genetic structure up to 100-m. The sampling scales differ, but mimicking this analysis, the structure in Maianthemum was only statistically significant up to 27-m. Considerable more insight can be gained when spatial statistics are augmented with AMOVA analyses to detect differentiation patterns. If high differentiation is present without spatial correlations then founding events of a single unrelated cohort are plausible (Torimaru et al. 2003). Highly differentiated patches as measured by Osx have been hypothesized to be due 31 to kin structuring in Silene dioica (Ingvarsson and Giles 1999) and Ilex leucoclada (Torimaru et al. 2003). As an aside, if clonemates were included, the O s t value would be inflated such that if unaccounted for, could lead to erroneous conclusions regarding gene flow (Reusch et al. 1999). The low O S T and correspondingly high molecular variance (77%) found within M. dUatatum patches are suggestive of the absence of kin-structuring or restricted gene flow among patches. Pair-wise relatedness within patches (rab=0.3) was higher than pair-wise comparisons between patches (rab=0.1) but the rab values had a wide range of -0. 07 to 0.8 (Table 2.2) which is indicative of the recruitment of both related and unrelated genets within patches. The highest diversity occurred within the largest patches (patches 2-4), with 85% of the molecular variance occurring within the patches, and a significant Spearman's rank diversity-size correlation. This positive correlation between diversity and patch size could be driven by seed-dispersers or pollinators being attracted by larger patches and or more genetically diverse mating opportunities. To differentiate among these hypotheses one would need to estimate visitation rates, conduct a yearly seedling census, use maternally inherited markers to detect family structure and estimate outcrossing rates. The fact that all seedlings were found within large patches provides some circumstantial support that seedling recruitment occurs more frequently in larger patches. Another reason for higher diversity in the larger patches could be that genets are congregating in high quality areas, but confirmation would require multi-year sampling to determine if new genets arise in a patch via seed or clonal dispersal. Additionally, in other species such as Solidago altissima genets in guerrilla species do become increasingly intermingled over time (Maddox et al. 1989) perhaps simply as a result of a scrambling habit. 2.5.5 Population level comparisons of genotypic and genetic variation Generalities which would be pulled from the transect data within the Burnaby Lake and Seymour populations is that areas of continuous cover will tend to be multi-genet patches. Nei's gene diversity measures were comparable among the populations and as discussed above were similarly high in comparison to other reported studies (Hamrick and Godt 1996). The sample sizes for the transects were much smaller than that for the Pacific Spirit population but this is compensated for by the large number of loci (Nei 1987). The absence of spatial structure is reflective of the data from Pacific Spirit since the spatial correlation was minimal beyond 30-m (Figs. 2.6 and 2.7) which was the sample interval used in the transects. This reinforces the fact that spatial genetic autocorrelation methods are scale-sensitive (Epperson 1993) and thus the sampling scale needs to be smaller than the scale of the correlation (Epperson and Li 1997). 32 Therefore, the M. dilatatum populations surveyed in this study were diverse, suggesting high connectivity among surrounding populations and sexual recruitment events occurring frequently enough to balance genet mortality. 2.6 Conclusions In summary, this chapter focused on the development of AFLP markers in Maianthemum dilatatum and the characterization of the clone structure at a finer scale within a single population. It was determined that M. dilatatum conforms to the clonal structure predictions for a guerrilla morphology, which is of dispersed clonemates with genetically heterogeneous patches. Patch margins only approximated genet margins in the small patches less than 10-m . There was a low level of spatial structure within the population which was possibly the product of both clonal growth and genet recruitment patterns. The populations of M. dilatatum studied seem to be genotypically and genetically diverse but to make any valid conclusion would require sampling additional populations with greater sampling effort. Although our sampling scale was fine compared to most, the ability to infer small-scale genetic processes is still limited. Despite the fact that patches are heterogeneous, the contribution of clonal growth to gene flow may operate at a finer level than the scale studied here. The next chapter will expand upon these findings and limitations by investigating the possible interference of clonal growth on pollination. 33 Table 2.1 Sequences of adaptors and primers used in this study Msel Adaptors 5 ' - G A C G A T G A G T C C T G A G 5 ' - T A C T C A G G A C T C A T EcoRI Adaptors 5 ' - A A T T G G T A C G C A G T C G T C 5 ' - A A C G A C G A C T G C G T A C C Msel Primers Pre-amplification 5 ' -GATGAGTCCTGAGTAA+CC Final Amplification 5 ' -GATGAGTCCTGAGTAA+CCAG 5 ' -GATGAGTCCTGAG' I AA+CCCA 5 ' -GATGAGTCCTGAGTAA+CCGG 5 ' -GATGAGTCCTGAGTAA+AT 5" -GATGAGTCCTGAG TAA+ATGT EcoRI Primers 5 ' -GACTGCGTACCAATTC+A 5 ' - C A C C G A C G T T G T A A A A C G A C G A C T G C G T A C C A A T T C + A G C 5 ' - C A C C G A C G l T G T A A A A C G A C G A C T G C G T A C C A A T T C + A G G 5 ' - C A C C G A C G T T G T A A A A C G A C G A C T G C G T A C C A A T T C + A G A 5 ' - G A C T G C G T A C C A A T I'C+C 5 ' - C A C C G A C G T T G T A A A A C G A C G A C T G C G T A C C A A T T C + C C C Table 2.2. Patch characteristics of the Pacific Spirit subpopulation Patch Patch Ramets No. of Genets Avg. r a b ±SE Range r a b Flowers Number Area (m2) Sampled Genets detected /m2 PI 9.00 4 1 Genet 1 0 P2 327.00 13 12 Genet 2-13 0.36 ±0.01 -0.01-0.79 0.95 P3 966.00 18 14 Genet 14-27 0.34 ±0.01 -0.04 -0.81 2.02 P4 1043.00 26 22 Genet 28-50 0.31 ±0.01 -0.07 -0.77 2.42 P5 48.98 3 1 Genet 51 0 P6 137.00 6 3 Genet 52-54 0.69 ± 0.07 0.58-0.85 1.21 P7 9.60 2 1 Genet 55 0.10 P8 327.67 7 3 Genet 56-58 0.46 ±0.12 0.28-0.83 1.73 P9 69.00 4 1 Genet 59 2.52 P10 70.94 4 2 Genet 60-61 0.033 1.68 Pl l 104.00 4 3 Genet 62-64 0.29 ±0.06 0.17-0.44 10.29 P12 21.00 3 1 Genet 65 0.43 P13 36.50 2 1 Genet 66 0.55 P14 37.90 4 1 Genet 67 1.64 P15 9.49 2 1 Genet 68 1.69 P16 1.00 1 1 Genet 69 0 P17 1.00 1 1 Genet 70 0 P18 6.00 1 1 Genet 71 0 P19 8.00 2 1 Genet 72 0.75 P20 3.00 2 1 Genet 73 0 P21 6.00 1 1 Genet 74 0 Table 2.3 Summary of the Lily of the Valley patch-level AMOVA analysis on all seven multi-genet patches with significance levels of p=0.001 which was based on 999 permutations. Source df SS MS Variance Components OST Prob. Among Patches 6 514.518 85.753 8.618 0.291 0.001 Within Patches 51 1066.792 20.917 20.917 34 30 20 h Q 10 0 •Sibling •Clonemate • Sibling- Clonemate Overlap 0.3 0.4 05 06 0.7 0.8 0.9 1.0 Jaccard Similarity Coefficients Figure 2.1. Distribution of Jaccard similarities for known siblings and clonemates. o O 12000 10000 3000 M H & 6000 4000 2000 -1 10 19 28 37 Number of Genets Removed 111 46 Figure 2.2. Decline of matrix incompatibility count with sequential genet removal for 50 putative genets based on 100 polymorphic loci. 35 2 0 0 0 1 5 0 0 + o S 1 0 0 0 4 -CU 5 0 0 + 0 •1 J l ITh-|T-i-i-r-t-[— 0 1 " 0 . 18 " 0 . 16 " 0 . 14 " 0 12 " 0 10 - 0 0 8 " 0 0 6 - 0 0 4 - 0 0 2 - -0 2 0 0 Pairwise Relatedness Coefficients within Pacific Spirit -d o O o CT Figure 2.3. Frequency distribution of pair-wise relatedness coefficients (rab) of sampled ramets within the Pacific Spirit subpopulation. 36 Figure 2.5. Mantel correlation test for pair-wise genetic and geographic distance for sampled ramets in the Pacific Spirit subpopulation (r2=0.1245, p=0.001). 16 25 35 45 55 63 71 79 87 95 106 122 150 187 Distance (m) Figure 2.6. Correlogram incorporating clonality, depicting r (correlation between genetic and geographical distance), 95% error bars about r with 95% CI given about the null hypothesis of r=0. 38 0.30 j -0.25 f -0.15 -<—1— !—•—•—^—i—>—:——i—i—:—;—i—'-• : — • — i — : — • — • • — : — : 1 16 25 35 45 55 63 71 79 87 95 106 122 150 187 Distance (m) Figure 2.7. Correlogram excluding clonality, depicting r (correlation between genetic and geographical distance), 95% error bars about r with 95% CI given about the null hypothesis r=0. 39 Chapter 3 The Role of Clonal Growth in Pollination Patterns 3.1 Introduction 3.1.1 Pollination ecology Pollination is an inherently inefficient process with pollen losses due to consumers, stochastic loss, asynchronous receptivity or stigma clogging with incompatible or heterospecific pollen (Inouye et al. 1994). Estimates suggest that <1 % of a donor's pollen will reach a conspecific stigma (Harder and Barrett 1996). Pollen mediated gene flow is highly dependent on spatial features such as density and dispersion. Pollinators respond to spatial features such that they are attracted by larger floral displays but within those displays will carry out non-random visitation generally within an inflorescence or between the nearest neighbouring floral ramets. Proximity-dependent mating success has been demonstrated several times in different breeding systems in monoecious (Nishihiro and Washitani 1998) and in particular dioecious species (Widen and Widen 1990). In animal-pollinated species, these patterns are driven exclusively by the propensity of pollinators to maximize foraging efficiency through minimizing flight distances (Richards 1986). When a pollinator picks up pollen, the load is only partially deposited on subsequent flowers visited. The extinction rate of a donor's load from the site of pickup is referred to as the carryover schedule (Richards 1986). In the bee-pollinated species Erythronium americanum, Clintonia borealis and Diervilla lonicera the carryover schedule revealed that 50 % of the load was deposited on the next flower visited with <1% remaining after the eighth visit (Thomson and Plowright 1980). Despite pollen carryover and documented cases of long-distance pollen flow (Broyles et al. 1994, Schulke and Waser 2001) pollen dispersal is generally strongly leptokurtic such that the most likely mating partners are the nearest neighbours, which for many clonal species would be clonemates (Handel 1985). 3.1.2 Clonality and pollination patterns When pollen transfer occurs between flowers within an inflorescence, this mode of pollination is termed geitonogamy, which is Greek for marriage between neighbours (Proctor et al. 1996). In both self-compatible and self-incompatible species, geitonogamy results in pollen discounting, which is the loss of pollen which could have been involved in outcrossing events 40 (Richards 1986). In self-compatible species, the self pollen will fertilize ovules and will constitute a selling event. In self-incompatible species, self pollen may reduce outcrossing opportunities by clogging the stigmatic surface and impeding the germination of compatible pollen grains (Barrett 2002). In aclonal plants, geitonogamy can occur between all open flowers occurring on the individual, but in clonal plants geitonogamy can also occur between clonemates which will be termed henceforth as within ramet geitonogamy and clonemate geitonogamy respectively. This additional level of geitonogamy has led to the prediction that higher rates of geitonogamous pollination should occur in clonal species (Silander 1985, Handel 1985). Clonemate proximity varies among clonal morphology types such that at the phalanx end of the gradient clonemates would tend to be aggregated with high potential for geitonogamy. As a phalanx species grows the interior ramets would become increasingly surrounded by clonemates and increasingly isolated from compatible mates. The predicted phalanx pollination pattern of increased geitonogamy with size has been demonstrated in Carexplatyphylla (Handel 1985), which sparked Handel's widely cited hypothesis of the positive correlation between genet size and the frequency of geitonogamy. At the guerrilla end of the gradient, clonemates would be more diffuse and intermingled among other genets and thus would be expected to be less prone to geitonogamous pollination as a function of genet size. Guerrilla and phalanx are opposite ends of a gradient of clonal spread coined by Lovett Doust (1981), and the use of these terms emphasizes the ecological diversity of clonal strategies which will influence the adaptiveness of particular mating strategies. Evidence of mating system differences among clonal morphologies was offered by Stebbins (1950), which in a comparison of 71 perennial species within seven genera in Graminaceae, found a tendency towards self-incompatibility in the guerrilla species (93%) and towards self-compatibility in the phalanx forms (77%). 3.1.3 Reproductive biology of the study species The flowering period for M. dilatatum occurs in late May to July, the duration seems to depend on the number of hot sunny days, where flowers open faster given hotter days. M dilatatum has a raceme with average of 28.56 ±1.06 (SE) flowers (pers. obs.). Flowers open sequentially, opening approximately 1-3 nodes at a time. Tepals turn from green to white 1-2 days prior to anthesis. Flowers are approximately 5-mm in diameter with four white tepals and with four longitudinally dehiscing anthers. Another unusual feature is the presence of septal 41 nectaries, which are nectaries within the pistil from which nectar is secreted through several pores on the surface (Kocyan and Endress 2001). The congener Maianthemum canadense is self-incompatible, (Worthen and Stiles 1986) and with Maianthemum tending towards a generalist pollination system (LaFrankie 1986) which was tested in this study through hand^pollinations and pollinator observations. There are typically four ovules which mature into 1-4 seeds. Seed maturation occurs by late November, signalled by the berries turning a translucent red colour. Seed dispersers of M. dilatatum are presumed to be avian such as thrushes but this was never directly observed. Observed seed consumers were squirrels. 3.1.4 Study objectives Handel's (1985) hypothesis needs additional empirical data from natural populations for a wider range of clonal morphologies, particularly in the guerrilla range of the spectrum. Clonal morphologies are structured according to a myriad of ecological pressures so it is important to measure the resulting mating patterns that arise from the actual spatial structure as found in natural populations. The objectives of this study were to: 1) Determine if the spatial arrangement of ramets within a guerrilla habit would result in clonemate aggregation. 2) Conduct pollinator observations to determine if visitation patterns are leptokurtic. 3) Determine the breeding system and pollination patterns of Maianthemum dilatatum. 3.2 Materials and methods 3.2.1 Pollination experiments Pollination experiments were done in three locations within Pacific Spirit Park and two locations in the Lower Seymour Conservation Reserve. The pollination treatment was a factorial experiment with two factors: hand-pollination and pollinator exclusion. The pollination factor had three levels, open (meaning unmanipulated) and the hand pollinations of selfed and outcrossed. Treatments were assigned according to random number assignment. Floral ramets selected for treatment were chosen such that treatment replicates were greater than 20m apart. Hand pollinations were done by collecting all anthers from the raceme of a single ramet from flowers of various ages, allowing unopened anthers to dehisce in tubes overnight and mixing the resulting pollen. The next day this pollen was then used to pollinate the donor ramet in selfed pollinations or in the outcrossed treatments, donor pollen would be used to pollinate a ramet from outside the donors immediate area >30m to avoid inadvertently using clonemates in 42 outcrossed pollinations. Pollinators were excluded with fine white mesh drawstring bags which were made to fit loosely over only the inflorescence. Fifty replicates of each treatment were completed but sample sizes were greatly reduced in the bagged individuals due to site disturbance and removal from analysis when pollinators were seen or suspected to have penetrated the bag. In summary, the six treatments along with remaining sample sizes were: Open bagged (n=17), Open unbagged («=45), Selfed bagged (n=20), Selfed unbagged (n=24), Outcrossed bagged («=17), Outcrossed unbagged (n=5\). Plants were monitored every 5 days throughout the flowering period until flowers had set fruit or withered. These treatments were used to test for stigmatic interference effects, self-compatibility and pollen limitation as described in greater detail below: Stigmatic interference experiment: The contribution of stigmatic clogging by geitonogamous pollination was tested by comparing bagged and unbagged outcrossed individuals. Self-interference would be supported if bagged individuals set higher seed than unbagged individuals. Self-compatibility: Self compatibility was be determined by comparing the seed set in outcrossed versus selfed individuals. Self-incompatibility will be demonstrated if hand-pollination using self pollen results in statistically lower seed set than outcross pollen. Pollen-limitation: Pollen-limitation was estimated by comparing unbagged outcrossed plants with unbagged open plants. Sixty-seven outcross pollinations were done with donors originating from outside the immediate area but from within 1 km to avoid possible outbreeding depression. To determine the flowering effort within the Pacific Spirit subpopulation, in 2002, all flowering ramets within all of the patches were counted. 3.2.2 Near-neighbour structure The study sites were in Pacific Spirit Park located in southwest Vancouver, British Columbia and Lower Seymour Conservation Reserve, North Vancouver. In these populations, nine plots were established wherein a focal floral ramet was randomly selected and the ten nearest flowering neighbours were sampled for genet identity. All ramets within the plot were measured and assigned coordinates to allow for distance measurements. 3.2.3 Pollinator behaviour Pollinator observations were conducted in 2002 and 2003, using lm diameter circular plots which were transiently established. Within these plots, all flowering ramets were mapped and labelled with black felt. Pollinators entering the plot had flight behaviour among the 43 labelled plants recorded. Some minute insects visit M. dilatatum, so any flights beyond 1 -m were ignored, because of the difficulty in following individual insects beyond this distance. Although this creates a bias towards short-distances (Proctor et al. 1996), this study was focusing on the short distance near-neighbour flights which is where most carry over pollen is believed to be deposited. Attempts were made with a fluorescent dye pollen analogue but were found to be inappropriate for dipteran pollinators as they invariably avoided marked plants. Additionally, beetles observed to pass through the dye did not pick up enough to be detected with a hand held black light. 3.2.4 Geitonogamy In 2002 self-incompatibility had not yet been confirmed in M. dilatatum, so manipulations were established to measure the contribution of between ramet geitonogamy. Even in a self-incompatible species emasculation effects can be used to compare self-interference due to ramet or clonemate level geitonogamy. This was done based on the approach of Eckert (2000) of using emasculation as a tool in geitonogamy studies where between-ramet geitonogamy rates can be estimated by the percentage of self seed set in open pollinated but emasculated plants. Any selfed seed in these maternal plants would necessarily come from other clonemates. Plants were emasculated by scooping out immature anthers from all unopened flowers within a raceme. Any open flowers were removed if present, or if dehisced anthers were found inside. Emasculated flowers opened in normal sequence and so this manipulation was assumed not to be detrimental to the female function. Unfortunately, the geitonogamy sample size was reduced to 17 due to unexpected construction which destroyed 28 plants before seed was fully matured. Nevertheless, 16 had no developing fruit as observed just prior to construction so could still be used in the emasculation data set. 3.2.5 DNA extraction protocol for leaf tissue The DNA extraction protocol was carried out as described in 2.2.5 3.2.6 DNA extraction protocol for endosperm The germination time for M. dilatatum seeds is reportedly nine months, but in preliminary trials low germination could be achieved using paper towel and tap water but there was almost complete damping off. In December 2002, a mass germination attempt was made with almost 400 berries where all seeds from a single berry were wrapped in autoclaved paper towel and placed in a microcentrifuge tube which was half-filled with sterile water. All manipulations and monthly waterings were done in the laminar flow hood. The germination protocol was unsuccessful in that it did not speed expected germination times with only 10 44 germinants by late July 2003. This difficulty led to the decision to use endosperm tissue along with the embedded embryo, which has been used successfully by Reusch (2000) for paternity analysis. Total genomic DNA was extracted from endosperm using a modified version of the extraction protocol offered by Kang et al. (1998). Seed coats were removed with a surgical blade and the endosperm with embryo finely sliced. The slices were placed with 600 u.1 of extraction buffer (200mM Tris-HCl (pH 8.0), 200mM NaCl, 25mM EDTA, 0.5% SDS, lOug/mL Proteinase K) and incubated at 37°C for 3 hours. Incubated slices were ground with a mini-pestle attached to a 12V drill in the buffer. Secondly, 600 u,l of extraction buffer (2% CTAB, 1.4M NaCl, 0.1M Tris-HCl, 20 mM EDTA, 2% p-mercaptoethanol, 1% PVP (polyvinylpyrrolidone Sigma MR 40, 000)) were added to the mixture and incubated at 65 °C for 1 hour. Two chloroform isoamyl alcohol 24:1 equal volume extractions followed. DNA in the supernatant was precipitated using 2/3 volume of isopropanol overnight at -20°C followed by a 30 minute spin at 4, 000 rpm at 4 °C. After discarding the isopropanol, the resulting DNA pellet was washed with 70%o ethanol. DNA was redissolved in 500(0,1 of sterile, distilled water and was then RNAse treated (10ug/mL) for 1 hour. Proteinase K treatment (10(ig/mL) followed for 60 minutes. Two equal volume extractions with phenol and chloroform:isoamyl followed, with a final 1:1 chloroform:isoamyl extraction. Final precipitation was done with two volumes of 100%o ethanol and NaCl to a final concentration of 0.15 M. DNA was spun down at 10, 000 rpm for 25 minutes followed by two washes with 70% Ethanol. DNA was resuspended in deionized water and was left overnight at 4°C before use in the AFLP protocol. Template DNA quantity was too low to be assessed for quality using the spectrophotometer. 3.2.7 AFLP protocol The AFLP protocol was carried out as described in 2.2.6 with the exception that seed DNA was estimated to be 5-10 ng/uf 3.3 Data analysis 3.3.1 Pollination experiments Berries of M. dilatatum hold a range of 1 -4 seeds, such that berry data alone could be misleading, so seed counts were used to measure the treatment effects. The seed set data did not conform to the assumptions of either an ANOVA nor a general linear model so a generalized linear model (GLZM) was needed. The GLZM model allows for a non-normal distributed unbalanced design. The Kruskal-Wallis test does not allow the calculation of interaction terms, 45 and it was felt that a bagging by pollination effect was plausible. To test for any trade-off between berry and seed production, a Spearman rank correlation was done. Flowering ramet census results were analysed using a GLZM model to test the relationship between patch size and total floral ramet count. 3.3.2 Near-neighbour structure Spatial patterning of the near-neighbour individuals was tested using both an index of aggregation (Clark and Evans 1954) and Thompson's test (Thompson 1956). The index of aggregation (R) is calculated as R=rA/rE where rA=the observed average pair-wise distance and rE=the average estimated pair-wise distance where significance testing is based on the standard normal deviate. The test is based on the expected distances between randomly positioned points as a function of density. 2 2 The Thompson's test is a two-tailed chi-square based test such that % =27ipS(rj ) and di=2nk where p=plot density, rj=distance to Mi near neighbour, k is the near-neighbour class and n is the sample size within this grouping. A boundary strip was not imposed during measurements but was implemented in sampling by using only known near-neighbour comparisons. Each measured ramet was given an allowable radius which corresponded to the closest distance to the edge of the plot. Only other ramets occurring within that allowable area would be used to calculate near-neighbour distances, ensuring that only true near-neighbour relationships are being depicted in Figures 3.6 and 3.7. The genotypic diversity of these plots was determined using AFLP clonal assignment criteria as explained in section 2.3.1 using Jaccard similarity coefficients with a pre-determined threshold. 3.3.3 Pollinator behaviour As depicted in Figure 3.3, 10% of the total 199 flights recorded were visitations where only a single ramet was visited and the pollinator left the study plot. For statistical reasons these flights were omitted from the flight distance and near neighbour analyses. The raw flight distances were square-root transformed and used in a general linear model (GLM) due to unbalanced sample sizes. Collective pollinator flight behaviours in terms of near-neighbour flights were tested using a x 2 against an expected distribution of equal visitation among all the ten nearest neighbours. 3.3.4 Geitonogamy The contribution of geitonogamous selfing was determined by the selfed seed set rates from the emasculated individuals. A GLZM was used to determine the differences in seed set 46 between open pollinated individuals which had been emasculated versus those who had not. If emasculation has no effect on seed set this would be suggestive that between ramet geitonogamy is more important than within ramet geitonogamy. If between clonemate geitonogamy is mostly responsible then clonality may have an impact on the outcrossing success of M. dilatatum. Selfing rates were determined using the paternity exclusion model to determine geitonogamous selfing patterns. All progeny were collected with the attached maternal tissue so only paternity is being inferred. Paternity exclusion is the appropriate model, given that there are a limited number of sires being tested and thus the value is not altered by the probability that the true sire is not among the potential sires being tested (Jones and Arden 2003). Exclusion simply involves summing the transmission mismatches among progeny and candidate fathers. The software used for this analysis was PROBMAX (Danzmann 1997). 3.4 Results 3.4.1 Pollination experiments The results of the pollination experiment are shown in Figure 3.1, where the GZLM model detected a significant interaction between the pollination and the exclusion treatment (X2=92.3991, pO.OOl). Excluding pollinators resulted in higher seed set with outcrossed individuals but seemed to reduce the seed set in open or selfed individuals. Within the different exclusion treatments, outcrossed individuals had higher seed than selfed and open pollinated individuals. Fruit set showed a statistically identical pattern which is expected given the high correlation of berries and seed produced within individuals (rs0.99, pO.OOl). Similarly, there was a high correlation (rsO.830, pO.OOl) between the number of berries and the number of seed per berry. A total of 6991 floral ramets were counted within the entire study area where the GLZM determined that patch area was significantly associated with flowering effort (Wald Stat X2=13380.55, pO.OOl). Flower counts for patches are reported in Table 2.2 where one outlier was patch 5 (Figure 2.4) which despite a moderate size produced no flowers. This study did not measure the successful fruit set as a function of the initial number of flowers. This was not felt to introduce bias to the pollination treatments given that the number of flowers within a raceme was relatively constant at 28.56 ±1.06 and any variability would be spread randomly across the treatments. 3.4.2 Near-neighbour structure The density and genotypic diversity of all nine measured plots are given in Table 3.1. The index of aggregation (R) resulted in only two plots showing a significant departure from 47 random but they both tended towards uniformity not aggregation. The Thompson's test resulted in all ten near-neighbour classes being almost exclusively random or uniform, with no aggregation being detected. AFLP analyses results were that five of the nine focal plots contained more than one genet (Table 3.1). In plots having multiple genets, representation among the ten nearest neighbours was very disproportionate. Figure 3.2 summarizes the proportion of individuals which encountered either a compatible mate or a clonemate in each of the ten neighbour classes measured. The density of the circle was not significantly correlated with the number of genets detected (rs=0.356, p=0.347). 3.4.3 Pollinator behaviour Field Observations: Maianlhemum dilatatum is a broad generalist with flower visitors ranging from (in descending frequency) dipterans, beetles, spiders and slugs. Dipterans were the most prevalent visitor to M. dilatatum flowers and as such are assumed to be the most important pollinators. Night observations were not conducted so the possibility of moth pollination cannot be excluded. Stigmatic exudates were often observed, occurring prior to anther dehiscence which could be indicative of a gametophytic incompatibility system and protogyny (Kearns and Inouye 1993). A total of 199 insects observations were made where 181 were within the plot and 18 were instances of an insect visiting a single ramet and flying beyond 1-m. As given in Table 3.2, flight distances were skewed to the left but a stronger pattern was seen in the average near-neighbour tendency of the visitors as seen in Figure 3.3. The visitation behaviour of each insect type is summarized in Table 3.2. Considering all pollinators the average flight distance was 17.4 cm +/-SE 0.821 and the average number of ramets visited was 2.3 +/-0.055. The GLM resulted in a non-significant effect of insect type on flight distance (F4 n6 =2.219, p=0.068), but a GLM test of insect type and number of ramets visited did find a significant effect ( F 4 17^ =4.923, p<0.001). Post-hoc Tukey comparisons revealed that both hoverfly and large beetles visited more ramets on average than small dipterans (p<0.05). Large beetles also visited more ramets than small beetles. The near-neighbour tendency showed a significant ( F 4 175 =3.202, p=0.014) effect with insect type. Flight distance and the near-neighbour tendency are correlated variables (r=0.56, p<0.001) and despite being statistically insignificant, flight distance showed a trend agreeing with the near neighbour tendency to differ among pollinators. Chi-square results (X2=215.36, p<0.05) confirmed that near-neighbour visitations were not spread equally among 48 the near-neighbour classes as can be seen in Figure 3.3. Near-neighbour transfers were much higher in the l s l to 3rd neighbour classes and lower in the 6th to 10th neighbour classes. 3.4.4 Geitonogamy Emasculating open pollinated individuals had no statistically significant effect (X =1.0991 p=0.2945) on seed set as depicted in Figure 3.4. Only ten individuals survived to set fruit for the geitonogamy measure and among those ten, the AFLP profiles from the endosperm were not of sufficient quality to infer parentage. Paternity exclusion tests determined that seeds were not selfed but also excluded the true mother. Therefore, it was concluded that the AFLP progeny data were unreliable, probably due to poor template quality. The profiles were not repeated because M. dilatatum appears to be largely self-incompatible, so selfed seed likely occurs at too low of a frequency to be included in such a small sample size such that repeating the reaction did not warrant the cost. 3.5 Discussion 5.5.7 Pollination experiments The significantly positive effect on seed set of outcross pollen and depressive effect of self pollen suggest that M. dilatatum is largely self-incompatible. The higher seed set found in the unbagged, outcross pollen-supplemented plants as compared to the open pollinated plants are indicative that compatible pollen is limiting. Determining incompatibility in plants can be difficult because of temporal variability in pollen viability and stigmatic receptivity such that a seemingly self-incompatible plant may actually have a mixed mating system (Vogler and Stephenson 2001). By pollinating the entire raceme, a range of flower ages and thus receptivities were used but there was no discernable pattern of fruit set within the raceme. Repeatedly visiting the raceme to pollinate individual flowers was felt to be too disruptive so an accepted source of error was that pollen may have arrived at the stigma in older flowers in all treatments. This is mentioned to discuss all sources of bias and because all individuals which set seed after self pollination were ones that had been left in the field. Maianthemum. dilatatum ramets can be very sensitive to transplanting and the impact of this stress on seed set was unknown which is why individuals were pollinated in situ. The pollination technique used in this study to determine self-incompatibility is a typical approach but differences in phenology of distant clonemates or cryptic self-fertility are factors which may allow for some selfing. Cryptic self-fertility can result from mixed pollen loads particularly in sporophytic compatibility systems (Bertin and Sullivan 1988). This phenomenon is due to the fact that the outcrossed pollen in the load triggers the stigmatic recognition to allow germination, such that 49 incompatible pollen will also germinate and have a reproductive chance. Therefore, despite the tests, it can only be concluded that there is low intra-ramet self-compatibility. Other studies of breeding systems in clonal plants such as Charpentier et al. (2000) found that in Scirpus maritimus pollination with near-neighbours resulted in statistically similar seed set to that of selfing. Therefore, it is plausible that the compatibility tests here denote a degree of self-incompatibility. The fact that M. dilatatum is self-incompatible does not negate its value as a study species for the cost of geitonogamy mediated through clonal growth. Upon comparison of the open and outcross treatments, M. dilatatum appears to be limiting in compatible pollen. As mentioned above, geitonogamy renders costs in self-incompatible species through pollen discounting and stigmatic interference. These costs are important to assess given that self-incompatibility is the prevalent breeding system found in the Angiosperms (Renner and Ricklefs 1995). Pollen supplementation results are also suggestive that seed set may be pollen limited given that in situ hand-pollinated individuals set significantly more seed than the open pollinated individuals (Fig. 3.1). If fruit set had been resource limited, such a high increase in seed set would not have occurred in outcross pollen supplemented plants (Fig. 3.1). Several species have been determined to be pollen limited despite the fact in some cases individuals had a significant stigmatic load yet little fruit set (Laverty and Plowright 1988, Schuster et al. 1993, Wilcock and Jennings 1999, Ushimaru and Kikuzawa 1999). Some of these studies reported negative correlations between clonal patch size and seed set (Laverty and Plowright 1988, Schuster et al. 1993) which would be suggestive of self-interference. These correlations would not be expected to be strong in M. dilatatum, because it was found that larger patches were associated with increased numbers of distinct genets (Fig. 2.4). The floral census gave expected patterns of more flowers concentrated in larger patches, which is believed to be because the larger patches are congregations of older genets. The other explanation would be that large patches are in high resource areas and or gap habitats. Light dependent reproductive effort is common in the shady understory and is suspected in M. dilatatum (van der Kamp pers. comm.). Regardless of the underlying mechanism, out of the number of sampled genets within Pacific Spirit subpopulation, seven genets could not contribute at all to the gene pool, reducing the effective population size within that flowering season. 3.5.2 Near-neighbour structure The ramet's eye view of the world sensu Turkington and Harper (1979), offers an interesting perspective in clonal plant ecology. The intraspecific interactions encountered by a 50 plant are primarily exchanged between neighbouring ramets (Harper 1977). The nature of these interactions is mediated by near-neighbour distances which are modified as a function of competition, clonal architecture (Crawley 1997) and environmental heterogeneity (Hutchings and Bradbury 1986). The mating consequences of clonality also partially operate at this scale which is why a floral ramet's eye view of geitonogamy was adopted in this study. For geitonogamy to be a cost associated with a clonal life history, one requirement would be that the neighbour structure encouraging geitonogamy must occur in natural populations. As seen in Figure 3.2, M. dilatatum genets do intermingle at a fine-scale, although single genet congregations were found in equal frequency. Within the sampling scheme, the scope was limited to the ten nearest neighbours so although genet representation was uneven, if all floral ramets within the plot had been sampled, this discrepancy might have been less pronounced. Intermingling at fine scale has been detected by Verburg et al. (2000) in Circaea lutetiana and Uvularia perfoliata (Kudoh et al. 1999). Despite some limitations, our sampling allows us to confirm that the near-neighbour spatial arrangement does create the spatial prerequisites for geitonogamy as well as outcrossing possibilities. Conversely, Ganger (1997) reported that infrequent flowering in Maianthemum canadense would result in low probability of the near-neighbour being the same genet. This prediction did not hold for M. dilatatum which often has profuse flowering displays and as shown has a high probability of near-neighbours being clonemates. As shown in Figure 3.3, the encounter probability of a compatible mate is relatively low but the more infrequent a genet is within a patch, the more likely a mate encounter will occur. This was demonstrated on a broader scale in the self-incompatible Calystegia collina where less frequent genets were more likely to set seed (Wolf et al. 2000). The genetic consequences of near-neighbour siring patterns have been shown through a simulation by Turner et al. (1982), where a strict near-neighbour siring scheme significantly lowered the heterozygosity within a population. Assuming the near-neighbour structure as shown in Figure 3.2, this pattern should occur to an even higher degree given that even the most diverse patches only had three genets and not the genetically distinct neighbours as in the simulation by Turner et al. (1982). 3.5.3 Pollinator behaviour Generalist pollination systems benefit from not depending on a single pollinator species whose abundance can vary substantially (Fishbein and Venable 1996) but this comes at the cost of higher pollen losses during transport and the risk of stigmatic clogging from heterospecific pollen (Wilcock and Neiland 2002). Maianthemum. dilatatum has a diverse assemblage of 51 pollinators which seem to vary in their visitation behaviour at a scale relevant to geitonogamy (Table 3.2). Pollinator flight distances between ramets were significantly skewed to the first three nearest neighbours. Yet, despite a strong nearest neighbour tendency for the first transfers in a foraging bout, most pollinators left the plot after two ramets and flew beyond 1 -m which in an older M. dilatatum patch should be a different genet. The number of near-neighbours between pollen donors and recipients were similar in Yuccafilamentosa where like in Figure 3.3, the majority of transfers were highly skewed (Marr et al. 2000). Knowledge of the behavioural responses of pollinators to spatial cues is almost exclusively restricted to bees, which although observed on other plant species within the study areas, were never seen on M. dilatatum flowers. Yet, effective short-term memory for pollen rewards as measured by constancy has been demonstrated in genera of anthophilous beetles (Englund 1993, Schlindwein and Wittmann 1997). In terms of pollinator service, beetle flight patterns among Viburnum opulus were not exclusively near-neighbour, ranging up to 18-m between ramets (Englund 1993). This scale exceeds my observations but if beetles behaved similarly for M. dilatatum based on genet structures described in Figure 2.2, the probability of clonemate interception is minimal and outcrossing almost certain. Dipteran pollination is not well characterized despite the overwhelmingly large number of plant species which they visit. In fact, there has been concern that the undocumented abundance of North American flies may result in population declines going unnoticed which would be detrimental for both insect and plant conservation (Kearns 2001). Although there were only trends but no significant effects of pollinator type on flight distance, pollinators were statistically different in near-neighbour tendencies and number of ramets visited within the plot. This finding agrees with previous studies, which have found different tendencies among pollinators in responses to density and floral traits (Thomson 2001). Pollinators tend to visit only a small proportion of available flowers (Lloyd and Schoen 1992), which has been attributed to satiation and exhausted reward supply (Snow et al. 1996). This behaviour was similarly seen in the M. dilatatum patches as the average number of flowers visited was only two (Fig. 3.3). As might be expected, the smaller bodied insects visited fewer flowers within the patch perhaps due to faster satiation and thus these insect types may contribute less to geitonogamy in microsites of low genet intermingling. Larger insects tended to visit more ramets, but would also travel to more distant neighbours along a visitation sequence, which would only contribute to geitonogamy more than the smaller bodied insects in low diversity plots. Maianthemum dilatatum's guerrilla tendencies create a situation where 52 pollinators vary in their contribution to geitonogamy depending on local levels of genet intermingling. The differences in contribution would depend on how the genets are arranged in relation to spatial behaviour of that pollinator type. The cost of these visitation patterns depends on the pollen carryover rate. Pollen carryover models generally assume a constant extinction of a donor's load, but Morris et al. (1994) suggest that the proportion of donor pollen lost actually decelerates. If the deposition decelerates during the visitation sequence then a higher proportion of the remaining pollen load will reach compatible conspecific stigmas, which lowers the pollen discounting. Given that only a few ramets within the 1-m plot were visited (Fig. 3.3) pollen carryover distances should be high in M. dilatatum, with much higher donor loads reaching compatible stigmas than if more ramets were visited within the plot. Therefore, based on behavioural observations, geitonogamy should occur at different rates among variously intermingled clonemates throughout the genet. Therefore, a generalist pollination system could be beneficial given the variable genetic structure of a guerrilla species. Conversely, it has been suggested that more dispersive morphologies should tend towards a specialist pollination system to avoid heterospecific pollen contamination (Feinsinger 1983). An important point is that observations of pollinator movements are only indirect measures of pollen mediated gene flow because pollen carryover will generally result in genes travelling beyond distances estimated from flight patterns alone (Thomson and Plowright 1980, Broyles and Wyatt 1991). The pollinator observations coupled with fine-scale genet structure in this study allow us to suggest that the opportunity for pollinator-mediated geitonogamy does exist but to varying degrees depending on local intermingling. Lastly, wind may dislodge pollen, such that clonemates in close proximity would suffer stigmatic clogging, even in the absence of pollinator visitation. 3.5.4 Geitonogamy Within ramet geitonogamy is a common dilemma to all plant species but geitonogamy between clonemates is exclusively pertinent to clonal species. In several self-compatible clonal species such as Iris versicolor (Back et al. 1996), Zoster a marina (Reusch 2001) and Decodon verticillatus (Eckert 2000), the majority of the selfing was due to clonemate geitonogamy. Within self-compatible species, the fitness cost of geitonogamy is through both pollen discounting and seed discounting which refers to the inbreeding depression in selfed progeny (Herlihy and Eckert 2002). Within self-incompatible species, geitonogamy and autogamy interfere with both female and male fitness through pollen discounting and self interference by 53 clogging the stigmatic surface with self pollen. Interference by self pollen in the germination of outcross pollen was found by in Polemonium viscosum (Galen et al. 1989), Asclepias exaltala (Broyles and Wyatt 1993), Campsis radicans (Bertin and Sullivan 1988) and Ipomopsis aggregata (Waser and Price 1991). This inhibiting ability of self pollen was even apparent on stigmas which were first pollinated by outcross pollen and in mixed loads (Broyles and Wyatt 1993). The pollen exclusion results coincide with those of Morse (1994), where hand pollinated individuals which were bagged set more seed than unbagged individuals which was similarly attributed to removing the interference effects from self-pollen. In this study, as shown in Figure 3.1, bagged, outcrossed individuals set significantly more seed than the outcrossed individuals which were unbagged. This seems counter-intuitive as the restriction of additional outcrossing opportunities should result in reduced relative fruit set in bagged individuals. Therefore, it is plausible that the outcrossed individuals which are bagged, are protected from the repressive effects of stigmatic clogging resulting from clonemate geitonogamy. A less significant factor may be pollen removal from the stigmas via post-depositional pollinivory by dipterans. Some support for the interference of clonemate geitonogamy can also be gained from the comparison of emasculated versus unemasculated open pollinated individuals (Fig. 3.8). There was no significant effect of emasculation, which implies that within ramet geitonogamy does not contribute in a major way to stigmatic interference. However, although not statistically significant there was a trend for higher seed set in the emasculated individuals which makes sense that within ramet geitonogamy would occur to some degree. In clonal species forming large monoclonal patches, geitonogamy could constitute a significant selective pressure; for example, towards dichogamy or dioecy (Thomson and Barrett 1981). The effectiveness of dichogamy for preventing geitonogamy between clonemates, requires that clonemates have synchrony in the phenology of the different sex phases. This clonemate-level synchrony has been reported in Aralia hispida (Thomson and Barrett 1981) but did not occur in Knautia arvensis (Vange 2002) nor in Alstroemeria aurea (Aizen and Basilio 1995). As genets increase in size, the phase synchrony between spatially distant clonemates is expected to decline, the rate of which depends on the developmental and environmental cues affecting phenology (Thomson and Barrett 1981). Spatially distant clonemates would be expected have a low probability of mating such that even if clonemates were asynchronous, there would be a low probability of pollen exchange between them. An obvious drawback to 54 clonemate synchrony would be missing opportunities to mate with other genets which may not be in complementary phases. Another approach to reducing geitonogamy would be the adoption of a guerrilla mode of growth. Unfortunately, clonal growth is essentially absent from evolutionary analyses so any association between mating patterns and clonal morphology is limited. Klimes et al. (1997) determined that despite the increased opportunity of selfing, clonal species are not any more likely to adopt a selfing mating system than aclonal species (Klimes et al. 1997). Selfing is only an effective means of reproductive assurance if the selfed progeny do not suffer from significant inbreeding depression and survive to reproductive maturity (Herlihy and Eckert 2002). Inbreeding depression manifested as smaller genet sizes in the self-compatible Zostera marina, which is intrinsically linked to genet survivorship (Hammerli and Reusch 2003). Furthermore, if size disparities among genets lead to unequal siring success, the effective population size would decline, leading to further genetic losses and inbreeding (Richards 1986, Ayres and Ryan 1997). Little is known about the genetic consequences of clonality but the potential for extreme genet longevities would suggest that reproductive assurance may be demographically and genetically unnecessary (Stebbins 1950, Bond 1994). In fact, any gains from reproductive assurance would be lost due to the fact that inbreeding depression may be comparatively more detrimental in clonal populations (Muirhead and Lande 1997) which potentially impacts future reproductive capacity (Morgan et al. 1997). 3.6 Conclusions This study aimed to test Handel's (1985) hypothesis that clonal growth has the potential to reduce outcrossing effectiveness. This was approached from a ramet's perspective such that the prerequisites of geitonogamy occurred in natural populations, namely that clonemates were the nearest neighbours and that pollinators did adopt a near-neighbour pattern of behaviour. Direct evidence of the depressive effects of geitonogamy was obtained from the pollination manipulations, in which individuals supplemented with outcross pollen, had enhanced seed set when pollinators were excluded. This result is counter-intuitive until one considers the possible role of geitonogamous pollination reducing female fitness. With respect to M. dilatatum, the greater dispersive power of a guerrilla habit increases the partner encounter likelihood which reduces, but does not eliminate the probability of geitonogamy. The clonal morphology and thus spatial arrangement of ramets has various constraints of which mating is only one. Certain levels of geitonogamy may need to be tolerated in order to 55 balance alternate ecological pressures (Reusch 2001). What was not demonstrated, is whether the cost of geitonogamy imposed by the clonal structure exceeds that arising from pollination inefficiency experienced by an aclonal species. In summary, it can be concluded that although the structure and pollinator behaviour should promote some level of geitonogamy in M. dilatatum, it is uncertain if the sexual interference and pollen discounting reduces reproductive potential to any level which would be relevant when scaled over a genet's lifespan. 56 Table 3.1. Near-neighbour genetic structure within 1-m diameter circular plot. Circle Number # Floral ramets # Genets 1 27 2 2 45 1 3 21 3 4 33 2 5 19 2 6 19 2 7 6 1 8 11 1 9 26 1 Table 3.2. Visitation behaviour of floral visitors in terms of flight distances (cm), number of ramets visited per 1-m plot, and near-neighbour transfer tendency given as weighted average ± SE Insect visitor type (sample size) Flight distance (cm) Number of ramets visited Neighbour transfer tendency Hoverfly (62) 16.38 ± 1.13 2.45 ±0.10 2.77 ± 0.229 Large Beetle (54) 20.99 ± 1.68 2.68 ±0.11 3.31 ±0.340 Small Beetle (20) 15.84 ± 3.18 2.25 ±0.12 2.35 ±0.371 Large Dipteran (11) 16.04 ±2.37 2.45 ±0.21 2.18 ±0.463 Small Dipteran (34) 15.35 ± 1.80 2.05 ± 0.04 1.91 ±0.195 Total Average (181) 17.48 ± 0.82 2.42 ± 0.05 2.69 ±0.146 57 tf ID a. aj CO o u SP 3 30 25 20 15 10 5 0 • • • • H I Outcrossed Bagged EBB Unbagged I I Open Pollination Treatment Selfed Figure 3 . 1 . Seed set per individual ramet (weighted mean ± SE) after pollination and pollinator exclusion treatments ( x 2 = 9 2 . 3 9 9 , pO.OOl). M N l NN2 NN3 NN4 NN5 NN6 NN7 NN8 NN9 NN10 • Encounters with different genet H Clonemate encounters Figure 3 . 2 . Proportion of the ten nearest neighbours which were clonemates (self encounter) or different genets (compatible mates). 58 70 60 £ 5 0 H | 40-H t 30 A 20-H 10 0 NN1 NN2 NN3 NN4 NN5 NN6 NN7 NN8 NN9 NN10 -=NN10 Ne ar-neighb oring transfer tendency Figure 3.3. The flight tendencies within the ten nearest neighbour classes during a pollinator's flight transfer between ramets. Unemasculated Emasculated Emasculation Treatment Figure 3.4. The effect of emasculation on the number of seed set per individual ramet (weighted mean ± SE) in open pollinated, unbagged plants (x2=1.0991, p=0.2945). 59 Future Directions In conclusion, clonality plays an important role in population structuring within Maianthemum dilatatum. The Pacific Spirit subpopulation had high genotypic diversity, driven by repeated seedling recruitment both within and outside of established patches. Patches became increasingly diverse, presumably through both recruitment and patch fusion occurring due to the guerrilla habit. Yet, given Maianthemum's patch forming ability, it is possible that an isolated population could quickly become genetically invariable, without a visible change in abundance. The majority of clonal species also rely on sexual reproduction, but the development of demographic and genetic models accounting for this has been limited. As it stands, very little is known about how a clonal habit would alter population and genetic dynamics within a species. Therefore, clonal species offer great potential for evolutionary and conservation research. Potential research areas are the evolutionary consequences of the spatial structure imposed by clonal growth. Data provided in this thesis demonstrated that the imposed close proximity of clonemates resulted in geitonogamy, which manifested in reduced reproductive output. Natural expansions on this work would be the scaling of reproductive costs to genet lifetimes. Unlike unitary species, clonal genets are potentially immortal such that fitness estimates need to be altered accordingly, in hopes of answering the question why not be clonal? Seemingly high costs may be irrelevant as compared to the gains derived from a clonal habit, but adaptations must arise to reduce the proximate influence of any disadvantages. Potential adaptations may be dioecy as previously mentioned, but others may be intra-clonal regulation of phenology or dispersal tendencies. Clonal plants are not sessile, and therefore a single genet may experience spatially and temporally variable selection, reducing the influence of stochastic genet mortality. Genet dispersal also introduces an interesting dilemma which is the interaction between disassociated clonemates, given that having many clonemates is intrinsically linked to genet survivorship. Furthermore, what generalities can be made regarding the effective population size (Ne) of clonal species? Clearly, the census size (N) is not valid, but it is necessary to develop some means of predicting genetic changes within clonal populations, for both academic and conservation reasons. It is imperative that clonality does not remain a side area of research, but becomes integrated into mainstream ecological thought and practice. In summary, this thesis aimed to contribute by first describing the genetic structure within a native clonal plant species; by understanding patch development within this species, a better estimate regarding the ratio of ramet to genets and thus N e can be made. Additionally, by focusing on geitonogamy which is one corollary of clonality, demonstrates that successful outcrossing rates cannot be accurately predicted based on the number of genets alone, but requires the integration of spatial aspects. 60 Literature Cited Abrahamson, W.G. 1980. Demography and vegetative reproduction. Demography and Evolution in Plant Populations (ed. O.T. Solbrig), pp. 89-106. University of California Press, Berkeley, California. Aizen, M.A. and Basilio, A. 1995. Within and among flower sex-phase distribution in Alstroemeria aurea (Alstroemeriaceae). Canadian Journal of Botany 73: 1986-1994. Allendorf, F.W. and Ryman, N. 2002. The role of genetics in population viability analysis. Population Viability Analysis, (eds. S.R. Beissinger andD.R. McCullough). University of Chicago Press. Chicago. Arens, P., Coops, H., Jansen, J. and Vosman, B. 1998. Molecular genetic analysis of black poplar (Populus nigra L.) along Dutch rivers. Molecular Ecology 7: 11-18. Auge, H., Neuffer, B., Erlinghagen, F., Grupe, R. and Brandl, R. 2001. Demographic and random amplified polymorphic DNA analyses reveal high levels of genetic diversity in a clonal violet. Molecular Ecology 10: 1811-1819 Avise, J. .1994. Molecular Markers, Natural History and Evolution. Chapman and Hall. New York. Ayres, D.R and Ryan, F.J. 1997. The clonal population structure of a rare endemic plant, Wyethia reticulata (Asteraceceae): allozyme and RAPD analysis. Molecular Ecology 6: 761-772 Back, A.J., Kron, P. and Stewart, SC. 1996. Phenological regulation of opportunities for within-inflorescence geitonogamy in the clonal species, Iris versicolor (Iridaceae). American Journal of Botany 83: 1033-1040 Baker, S.S., Rugh, C L . and Kamalay, J.C 1990. RNA and DNA isolation from recalcitrant plant tissue. Biotechniques 9:268-272 Balloux, F., Lehmannc, L. and de Meeus, T. 2003. The population genetics of clonal and partially clonal diploids. Genetics 164: 1635-1644 Barbour, M.G., Burk, J.H. and Pitts, W.D. 1987: Terrestrial Plant Ecology Benjamin/Cummings, Menlo Park CA Barrett, S. C. H., Harder, L. D. and Worley, A. C. 1996. The comparative biology of pollination and mating in flowering plants. Philosophical Transactions of the Royal Society of London Series B 351: 1271-1280 Barrett, S.C.H. 2002. Sexual interference of the floral kind. Heredity 88: 154-159 Bell, A.D. 1991. Plant Form, An Illustrated Guide to Flowering Plant Morphology. Oxford University Press, New York. Bennett, M.D. and Leitch, l.J. 2003. Angiosperm DNA C-values database (release 4.0, Jan. 2003) Bertin, R.I. and Sullivan, M. 1988. Pollen Interference and cryptic self-fertility in Campsis radicans. American Journal of Botany 75: 1140-1147 Bharathan, G., Lambert, G. and Galbraith, D.W. 1994. Nuclear DNA content of 61 monocotyledons and related taxa. American Journal of Botany 81: 381-386 Bierzychudek, P. 1982. Life histories and demography of shade-tolerant temperate forest herbs: a review. New Phytologist, 90: 757-776. Bold, H.C., Alexopoulos, C.J. and Delevoryas, T. 1987. Morphology of Plants and Fungi 5th ed. Harper Collins Publishers, New York. Bond, W.J. 1994. Do mutualisms matter - Assessing the impact of pollinator and disperser disruption on plant extinction. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 344: 83-90 Bonner, J. T., 1988. The Evolution of Complexity by Means of Natural Selection. Princeton University Press. Princeton, New Jersey: Brown, J. K. M. 1996. The choice of molecular marker methods for population genetic studies of plant pathogens. New Phytologist 133:183-195. Broyles, S. B. and Wyatt, R. 1991. Effective pollen dispersal in a natural population of Asclepias exaltata, the influence of pollinator behavior, genetic similarity, and mating success. American Naturalist 1239-1249 Broyles, S.B. and Wyatt, R. 1993. The consequences of self-pollination in Ascelpias exaltata, a self-incompatible milkweed. American Journal of Botany 80: 41-44 Broyles, S.B., Schnabel, A. and Wyatt, R. 1994. Evidence for long-distance pollen dispersal in Milkweeds (Asclepias exaltata). Evolution 48: 1032-1040 Burrows, C.J. 1990. Processes of Vegetation Change. Unwin Hyman, London Caswell, H. 1985. The evolutionary demography of clonal reproduction. Population Biology and Evolution of Clonal Organisms (eds. J. B. C. Jackson, L. W. Buss and R. E. Cook), pp. 187-244. Yale University Press, New Haven. Charpentier, A., Grillas, P. and Thompson, J.D. 2000. The effects of population size limitation on fecundity in mosaic populations of the clonal macrophyte Scirpus maritimus (Cyperaceae). American Journal of Botany 87: 502-507 Cheplick, G.P. 1995. Life history trade-offs in Amphibromus scabrivalvis (Poaceae): allocation to clonal growth, storage, and cleistogamous reproduction. American Journal of Botany 82: 621-629. Chung, M.G. and Epperson, B.K. 2000. Clonal and spatial genetic structure in Eurya emarginata (Theaceae). Heredity 84: 170-177 Clark, P. J. and Evans, F.C. 1954. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology 35: 445-453. Cody, M L . 2000. Slow-motion population dynamics in Mojave Desert perennial plants. Journal of Vegetation Science 11: 351-358 Collado-Vides, L. 2001. Clonal architecture in marine macroalgae: ecological and evolutionary perspectives. Evolutionary Ecology 15: 531-545 Cook, R.E. 1983. Clonal plant populations. American Scientist 71: 244-253. Cousens, M.I. 1988. Reproductive Strategies of Pteridophytes. Plant Reproductive Biology, Patterns and Strategies, (eds. J. Lovett Doust and L. Lovett Doust), pp 307-325. 62 Oxford University Press, Oxford. Crawley, M.J. 1997. Plant Ecology. 2nd edition. Blackwell Science, Oxford. D'Hertefeldt, T. and van der Putten, W.H. 1998. Physiological integration of the clonal plant Carex arenaria and its response to soil-borne pathogens. Oikos 81: 229-237 Damman, H. and Cain, M.L. 1998. Population growth and viability analyses of the clonal woodland herb, Asarum canadense. Journal of Ecology 86: 13-26. Danzmann, R.G. 1997. PROBMAX: A computer program for assigning unknown parentage in pedigree analysis from known genotypic pools of parents and progeny. Journal of Heredity 88: 333. Dassler, C L and Farrar, D.R. 1997. Significance of form in fern gametophytes: Clonal, gemmiferous gametophytes of Callistopteris baueriana. International Journal of Plant Sciences 158: 622-639 Del Tredici, P, Ling,H. and Yang, G. 1992. The Ginkgos of Tian Mu Shan. Conservation Biology 6: 202-209. Del Tredici, P. 1992. Natural regeneration of Ginkgo biloba from downward growing cotyledonary buds (basal chichi). American Journal of Botany 79: 522-530 Dittmer, H.J. 1964. Phylogeny and Form in the Plant Kingdom. Van Nostrand Company Inc, New York. Do, N. and Adams, R.P. 1991. A simple but powerful technique for removing polysaccharide contaminants from DNA. Biofeedback 10: 162-166 Doligez, A., Baril, C. and Joly, H.I. 1998. Fine-scale spatial genetic structure with nonuniform distribution of individuals. Genetics 148: 905-919 Douglas, G.W., Straley, G.B. and Meidinger, D.V. 1997. 77ze Vascular Plants of British Columbia. Special report series British Columbia. Ministry of Forests. Victoria, BC. Douhovnikoff, V. and Dodd, R.S. 2003. Intra-clonal variation and a similarity threshold for identification of clones: Application to Salix exigua using AFLP molecular markers. Theoretical and Applied Genetics 106: 1307-1315. During, H.J. 1990. Clonal growth patterns among bryophytes. Clonal Growth in Plants: Regulation and Function (eds. J. van Groenendael and H. de Kroon.), pp. 153-176. SPB Academic Publishing, The Hague Eckert, C. G., Keiko, L., Bronson, K., Corradini, P. and Bruneau, A. 2003. Population genetic consequences of extreme variation in sexual and clonal reproduction in an aquatic plant. Molecular Ecology 12: 331-344. Eckert, C G . 2000. Contributions of autogamy and geitonogamy to self-fertilization in a mass-flowering, clonal plant. Ecology 81: 532-542 Eckert, C G . and Barrett, S.C.H. 1993. Clonal reproduction and patterns of genotypic diversity in Decodon verticillatus (Lythraceae). American Journal of Botany 80: 1175-1182 Ehrlen, J. and Lehtila, K. 2002. How perennial are perennial plants? Oikos 98: 308-322. Ellstrand, N.C. and Roose, M.L. 1987. Patterns of genotypic diversity in clonal plant 63 species. American Journal of Botany 74: 123-131. Englund, R. 1993. Movement patterns of Cetonia Beetles (Scarabaeidae) among flowering Viburnum opulus (Caprifoliaceae) - option for long-distance pollen dispersal in a temperate shrub. Oecologia 94: 295-302 Ennos, R. A. 2001. Inferences about spatial processes in plant populations from the analysis of molecular markers. Integrating Ecology and Evolution in a Spatial Context (eds. J. Silvertown and J. Antonovics), pp. 45-71. Blackwell Science, Oxford. Epperson, B. K. 1993. Recent advances in correlation studies of spatial patterns of genetic variation. Evolutionary Biology.27:95-155 Epperson, B.K. and Li, T. 1997. Gene dispersal and genetic structure. Evolution 51: 672-681 Eriksson, O. 1989. Seedling dynamics and life histories in clonal plants. Oikos 55:231-238 Eriksson, 0. 1993. Dynamics of genets in clonal plants. Trends in Ecology and Evolution 8: 313-316. Eriksson, O. 1994. Stochastic population dynamics of clonal plants: Numerical experiments with ramet and genet models. Ecological Research 9: 257-268 Eriksson, O. 1997. Colonization dynamics and relative abundance of three plant species (Antennaria dioica, Hieracium pilose/la and Hypochoeris maculata) in dry semi-natural grasslands. Ecography 20:559-568 Eriksson, O. and Froborg, H. 1996. "Windows of opportunity" for recruitment in long-lived clonal plants: Experimental studies of seedling establishment in Vaccinium shrubs. Canadian Journal of Botany 74: 1369-1374 Excoffier, L, Smouse, P.E. and Quattro, J.M. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mtDNA restriction data. Genetics 131: 479-491 Fager, E.W. 1972. Diversity — a sampling study. American Naturalist 106: 293-310. Falik, O., Reides, P., Gersani, M., and Novoplansky, A. 2003. Self/non-self discrimination in roots. Journal of Ecology 91: 525-531. Farrar, J.L. 1995. Trees of Canada. Fitzhenry and Whiteside Limited, Markham, Ontario Feinsinger, P. 1983. Coevolution and pollination. Coevolution (eds. D.J. Futuyama, and M. Slatkin) pp. 282-331, Sinauer Associates, Sunderland, Massachusetts Fishbein, M. and Venable, D.L. 1996. Diversity and temporal change in the effective pollinators of Asclepias tuberosa. Ecology 77: 1061-1073 Galen, C.T., Gregory, T., and Galloway, L.F. 1989. Costs of self-pollination in a self-incompatible plant, Polemonium viscosum. American Journal of Botany 76: 1675-1680 Ganger, M.T. 1997. The influence of pollen addition and ramet isolation on current sexual reproduction in a clonal herb. Oecologia 110: 231-236 Garner, T.W.J. 2002. Genome size and microsatellites: the effect of nuclear size on amplification potential. Genome 45: 212-215 Gaudeul, M., Taberlet, P. and Till-Bottraud, I. 2000. Genetic diversity in an endangered 64 alpine plant, Eryngium alpinum L. (Apiaceae), inferred from amplified fragment length polymorphism markers. Molecular Ecology 9: 1625-1637 Gifford, E.M and Foster, A.S. 1988. Morphology and Evolution of Vascular Plants. 3rd ed. W H Freeman and Co. New York Gough, L., Goldberg, D.E., Hershock, C , Pauliukonis, N, and Petru, M. 2001. Investigating the community consequences of competition among clonal plants. Evolutionary Ecology 15: 547-563 Halle, F., Oldeman, R. A. A. and Tomlinson, P.B. 1978. Tropical Trees andForests-An Architectural Analysis. Springer-Verlag New York Hammerli, A. and Reusch, T.B.H. 2003. Inbreeding depression influences genet size distribution in a marine angiosperm. Molecular Ecology 12: 619-629 Hamrick, J.L. and Godt, M.J.W. 1996. Effect of life history traits on genetic diversity in plant species. Philosophical Transactions of the Royal Society of London series B-Biological Sciences 351: 1291-1298 Handel, S.N. 1985. The intrusion of clonal growth patterns on plant breeding systems. American Naturalist 125: 367 384. Harder, L. D. and Barrett, S. C. H. 1996. Pollen dispersal and mating patterns in animal-pollinated plants. Floral biology: Studies on Floral Evolution in Animal Pollinated Plants. (eds. D. G. Lloyd and S. C. H. Barrett), pp. 140-190. Chapman and Hall, New York. Hardy, O J. and Vekemans, X. 2002. SPAGeDi: a versatile computer program to analyze spatial genetic structure at the individual or population levels. Molecular Ecology Notes 2: 618-620 Hardy, O.J. 2003. Estimation of pair-wise relatedness between individuals and characterization of isolation-by-distance processes using dominant genetic markers. Molecular Ecology 12: 1577-1588 Harper, J.L and White, J. 1974. The demography of plants. Annual Review of Ecology and Systematics 5: 419—163 Harper, J.L. 1977. Population Biology of Plants. Academic Press, London Harper, J.L. 1980. Plant demography and ecological theory. Oikos 35: 244-253 Hay, M.E. 1981. The functional morphology of turf forming seaweeds. Ecology 62: 739-750 Herlihy, CR. and Eckert, C G . 2002. Genetic cost of reproductive assurance in a self-fertilizing plant. Nature 416: 320-323 Heuertz, M., Vekemans, X., Hausman, J.F., Palada, M. and Hardy, O.J. 2003. Estimating seed vs. pollen dispersal from spatial genetic structure in the common ash. Molecular Ecology 12: 2483-2495 Heywood, J.S. 1991. Spatial analysis of genetic variation in plant populations. Annual Review of Ecology and Systematics 22:335-355 Hitchcock, C L . and Cronquist, A. 1997. Flora of the Pacific Northwest. University of Washington Press, Seattle. 65 Holderegger, R., Stehlik, I. and Schneller, J.J. 1998. Estimation of the relative importance of sexual and vegetative reproduction in the clonal woodland herb Anemone nemorosa. Oecologia 177: 105-107 Holzapfel, C. and Alpert, P. 2003. Root cooperation in a clonal plant: connected strawberries segregate roots. Oecologia 134: 72-77 How, S.T., Abrahamson, W.G. and Zivitz, M.J. 1994. Disintegration of clonal connections in Solidago altissima (Compositae). Bulletin of the Torrey Botanical 121: 338-344 Huff, D.R., Peakall, R. and Smouse, P.E. 1993. RAPD variation within and among natural populations of outcrossing populations of outcrossing buffalograss (Buchloe dactyoides (Nutt.) Engelm.). Theoretical and Applied Genetics 86: 927-934. Hutchings, M.J. and Bradbury, I.K. 1986. Ecological perspectives on clonal perennial herbs. Bioscience 36: 178-182 Inghe, O. and Tamm, CO. 1985. Survival and flowering of perennial herbs. IV. The behaviour of Hepatica nobilis and Sanicula europaea on permanent plots during 1943 and 1981. Oikos 45:400-420. Ingvarsson, P. K. and Giles, B. E. 1999. Kin-structured colonization and small-scale genetic differentiation in SHene dioica. Evolution 53: 605-611. Innes, D. J. 1987. The genetic structure of asexually reproducing Enteromorph lima (Ulvales, Chlorophyta) in Long Island Sound. Marine Biology 94: 459-467 Inouye, D.W., Gill, D.E., Dudash, M.R. and Fenster, CB. 1994. A model and lexicon for pollen fate. American Journal of Botany 81: 1517-1530. Isabel, N., Beaulieu, J., Theriault, P., and Bousquet, J. 1999. Direct evidence for biased gene diversity estimates from dominant random amplified polymorphic DNA (RAPD) fingerprints. Molecular Ecology 8: 477-484 Jacobson, K.M and Lester, E. 2003. A first assessment of genetic variation in Welwitschia mirabilis Hook. Journal of Heredity 94: 212-217 Janzen, D.H. 1977. What are dandelions and aphids? American Naturalist 111: 586-589 Jobes, D.V., Hurley, D.L. and Thien, L.E. 1995. Plant DNA isolation: a method to efficiently remove polyphenolics, polysaccharides, and RNA. Taxon 44: 379 - 385. Johansson, M. E., and Nilsson, C. 1993. Hydrochory, population dynamics and distribution of the clonal aquatic plant Ranunculus lingua. Journal of Ecology 81: 81-91. Johnson, S. D., and Steiner, K. M. 2000. Generalization versus specialization in plant pollination systems. Trends in Ecology and Evolution 15: 140-143. Jones, A.G. and Arden, W.R. 2003. Methods of paternity analysis in natural populations. Molecular Ecology 12:2511-2523 Jones, D.L. 1987. Encyclopedia of Ferns. Lothian Publishing Company Ltd, Melbourne. Jonsdottir, I. S. and Callaghan, T. V. 1989. Localized defoliation stress and the movement of C 1 4 photoassimilates between tillers of Carex bigelowii. New Phytologist 114:419-428. Kang, H.W., Cho, Y.G., Yoon, U.H. and Eum, M.Y. 1998. A rapid DNA extraction for RFLP and PCR analysis from a single dry seed. Plant Molecular Biology Reporter 16: 1-9 66 Kawano, S., Suzuki, M. and Kojima, S. 1971. Biosystematic studies on Maianthemum (Liliaceae-Polygonatae), 5. Variation in gross morphology, karyology, and ecology of North American populations of M.dilatatum sensu lato. Botanical Magazine 84:299-318 Kearns, C. A. 2001. North American dipteran pollinators: assessing their value and conservation status. Conservation Ecology 5: 5 URL: Kearns, C. A. and inouye, D. W. 1993, Techniques for Pollination Biologists. University Press of Colorado, Colorado. Kelber, K.P. and van Konijnenburg-van Cittert, J.H.A. 1998. Equisetites arenaceus from the Upper Triassic of Germany with evidence for reproductive strategies. Review of Paleobotany and Palynology 100: 1-26 Kelly, C.K. 1995. Thoughts on clonal integration -facing the evolutionary context. Evolutionary Ecology 9: 575-585 Kemperman, J.A and Barnes, B.V. 1976. Clone size in American Aspens. Canadian Journal of Botany 54: 2603-2607. Kinsman, S. 1990. Regeneration by fragmentation in tropical montane forest shrubs: American Journal of Botany 77: 1626-1633. Klimes, L., Klimesova, J., Hendricks, R and van Groenendael, J. 1997. Clonal plant architecture: A comparative analysis of form and function. The Ecology and Evolution of Clonal Plants (eds. H de Kroon and J. van Groenendael), pp. 1-29. Backhuys Publishers, Leiden Kocyan, A., and Endress, P.K. 2001. Floral structure and development and systematic aspects of some 'lower' Asparagales. Plant Systematics and Evolution 229: 187-216 Koonjul, P.K., Brandt, W.F., Farrant, J.M. and Lindsey, G.G. 1999. Inclusion of polyvinylpyrrolidone in the polymerase chain reaction reverses the inhibitory effects of polyphenolic contamination of RNA. Nucleic Acids Research 27: 915-916 Kreher, S.A., Fore, S.A. and Collins, B.S. 2000. Genetic variation within and among patches of the clonal species, Vaccinium slamineum L. Molecular Ecology 9: 1247-1252 Kubitzki, K. 1990. Ephedraceae. The Families of Genera of Vascular Plants Vol 1 Pteridophytes and Gymnosperms (ed. K. Kubitzki), pp. 379-382. Springer-Verlag, New York. Kudoh, H., Shibaike, H., Takasu, H., Whigham, D. and Kawano, S. 1999. Genet structure and determinants of clonal structure in a temperate deciduous woodland herb, Uvular ia perfoliata. Journal of Ecology 87: 244-257 Laaka-Lindberg, S., Korpelainen, H. and Pohjamo, M. 2003. Dispersal of asexual propagules in bryophytes. Journal of The Hattori Botanical Laboratory 93: 319-330 LaFrankie, J. V. 1986. Morphology and taxonomy of the new world species of Maianthemum (Liliaceae). Journal of the Arnold Arboretum 67: 371-439 Laverty, T.M. and Plowright, R.C. 1988. Fruit and seed set in mayapple {Podophyllum peltatum): influence of intraspecific factors and local enhancement near Pedicularis canadensis. Canandian Journal of Botany 66: 173-178 67 Lazo, M.L. and Chapman, A.R.O. 1998. Components of crowding in a modular seaweed: sorting through the contradictions. Marine Ecology Progress Series 174: 257-268 Leakey, R.R.B. 1981. Adaptive biology of vegetatively regenerating weeds: Advances in Applied Biology 6: 57-90. Lewontin, R.C. 1972. The apportionment of human diversity. Evolutionary Biology 6: 381 394 Lezberg, A.L., Antos, J.A. and Halpern, C B . 1999. Belowground traits of herbaceous species in young coniferous forests of the Olympic Peninsula, Washington. Canadian Journal of Botany 77: 936-943 Lloyd, D. G. and Schoen, D. J. 1992. Self- and cross-fertilization in plants. I. Functional dimensions. International Journal of Plant Sciences 153:358-369. Lobban, CS. and Harrison, P.J. 1994. Seaweed Ecology and Physiology. Cambridge University Press, New York. Lobban, CS. and Wynne, M.J. 1981. Biology of Seaweeds. Blackwell Scientific Publications, Oxford. Loveless, M. D. and Hamrick, J. L. 1984. Ecological determinants of genetic structure in plant populations. Annual Review of Ecology and Systematics 15: 65-95 Lovett Doust, L. 1981. Population dynamics and local specialization in a clonal perennial (Ranunculus repens). 1. The dynamic of ramets in contrasting habitats. Journal of Ecology 69: 743-755 Lynch, M and Walsh, B.G. 1998. Genetics and Analysis of Quantitative Traits. Sinauer Associates Inc. Sunderland, Massachusetts. Lushai, G., Loxdale, H.D. and Allen, J.A. 2003. The dynamic clonal genome and its adaptive potential. Biological Journal of the Linnean Society 79: 193-208 Lynch, M. and Milligan, B.G. 1994. Analysis of population genetic structure with RAPD markers. Molecular Ecology 3: 91-99 Maddox, G. D., R. E. Cook, Wimberger, P. H. and Gardescu, S. 1989. Clone structure in four Solidago altissima (Asteraceae) populations: rhizome connections within genotypes. American Journal of Botany 76:318-326. Marr, D.L., Leebens-Mack, J., Elms, L. and Pellmyr, O. 2000. Pollen dispersal in Yucca fdamentosa (Agavaceae): the paradox of self-pollination behavior by Tegeticula yuccasella (Prodoxidae). American Journal of Botany 87: 670-677 Marshall, C. and Price, E.A.C. 1997. Sectoriality and its implications for physiological integration. The Ecology and Evolution of Clonal Plants, (eds. H. de Kroon and J. van Groenendael), pp. 79-108. Backhuys Publishers, Leiden . McLellan A.J, Prati, D, Kaltz, O., and Schmid, B. 1997. Structure and analysis of phenotypic and genetic variation in clonal plants. The Ecology and Evolution of Clonal Plants, (eds. H. de Kroon and J. van Groenendael), pp. 185-210. Backhuys Publishers, Leiden. McNeilly, T and Roose, M. L. 1984. The distribution of perennial ryegrass genotypes in swards. New Phytologist 98: 503-513 68 Meagher, T. R., and Vassiliadis, C. 2001. Spatial geometry determines gene flow in plant population. Genes in the Environment (eds. R. Hails, J. Beringer, and H. C. J. Godfray), pp.76-90. Blackwell Publishing, Oxford. Mes, T.H.M. 1998. Character compatibility of molecular markers to distinguish asexual and sexual reproduction. Molecular Ecology 7: 1719-1727 Mishler, B. 1988. Bryophyte Reproductive Strategies. Plant Reproductive Ecology (eds. J. Lovett Doust and L. Lovett Doust), pp. 285-306. Oxford University Press, New York. Mitton, J.B. and Grant, M.C. 1996. Genetic variation and natural history of quaking aspen. Bioscience 46:25-31. Mogie, M. and Hutchings, M.J. 1990. Phylogeny, ontogeny and clonal growth in vascular plants. Clonal Growth in Plants: Regulation and Function, (eds. J. van Groenendael and H. de Kroon) pp. 3-22. SPB Academic Publishing, The Hauge. Montalvo, A.M., Conard, S.G., Conkle, M.T. and Hodgskiss, P.D. 1997. Population structure, genetic diversity, and clone formation in Quercus chrysolepis (Fagaceae). American Journal of Botany 84: 1553-1564 Morgan, M. T., Schoen, D. J. and. Bataillon, T. M. 1997. The evolution of self-fertilization in perennials. American Naturalist 150: 618-638. Morris, W.F., Price, M.V., Waser, N.M., Thomson, J.D., Thomson, B. and Stratton, D.A. 1994. Systematic increase in pollen carryover and its consequences for geitonogamy in plant populations. Oikos 71: 431-440 Morse, D.H. 1994. The role of self-pollen in the female reproductive success of common milkweed (Asclepiassyriaca, Asclepiadaceae) American Journal of Botany 81: 322-330 Mueller, U. G. and Wolfenbarger, L. L. 1999. AFLP genotyping and fingerprinting. Trends in Ecology and Evolution 14:389-394. Muirhead, C A . and Lande, R. 1997. Inbreeding depression under joint selfing, outcrossing, and asexuality. Evolution 51: 1409-1415 Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University Press, New York. Newell, S.J. 1985. Reproduction in a Natural Population of Cycads (Zamia pumila) in Puerto Rico. Bulletin of the Torrey Botanical Club 110: 464-473 Nishihiro, J. and Washitani, I. 1998. Effect of spatial structure on pollination and seed set of a clonal distylous plant, Persicaria japonica. (Polygonaceae) Journal of Plant Research 111: 547-555 Norstog, K.J. and Nicholls, T.J. 1997. The Biology of the Cycads. Cornell University Press, New York. Parks, J.C. and Werth, CR. 1993. A study of spatial features of clones in a population of bracken fern, Pteridium aquilinum (Dennstaedtiaceae). American Journal of Botany 80: 537-544 Peakall, R. and Smouse, P. E. 2001. GenAlEx V5: Genetic Analysis in Excel. Population genetic software for teaching and research. Australian National University, Canberra, Australia. Peltzer, D.A. 2002. Does clonal integration improve competitive ability? A test using 69 aspen (Populus tremuloides) invasion into prairie. American Journal of Botany 89: 494-499 Pielou, E.C. 1969. An Introduction to Mathematical Ecology. Wiley-lnterscience, New York Piqueras, J. and Klimes, L. 1998. Demography and modelling of clonal fragments in the pseudoannual plant Trientalis europaea L. Plant Ecology 14: 213-217. Pitelka, L. F. and Ashmun, J. W. 1985. Physiology and integration of ramets in clonal plants. Population Biology and Evolution of Clonal Organisms, (eds. J. B. C. Jackson, L.W. Buss and R.E. Cook), pp. 399^ 135. Yale University Press, New Haven. Porebski, S., Bailey, L.G. and Baum, B.R. 1997. Modification of a CTAB DNA extraction protocol for mature strawberry leaves (Fragaria spp.) Containing high polysaccharide and polyphenol components. Plant Molecular Biology Reporter 15: 8-15. Powell, W., Morgante, M., Andre, C , Hanafey, M., Vogel, J., Tingey, S. and Rafalski A. 1996. The comparison of RFLP, RAPD, AFLP and SSR (microsatellite) markers for germplasm analysis. Molecular Breeding 2: 225-238. Proctor, M., Yeo, P. and Lack, A. 1996. The Natural History of Pollination. Timber Press Ltd Oregon Pysek, P. 1997. Clonality and plant invasions: can a trait make a difference? The ecology and evolution of clonal plants (eds. H. de Kroon and J. van Groenendael), pp. 405^427, Backhuys Publishers, Leiden. Renner, S.S., and Ricklefs, R.E. 1995. Dioecy and its correlates in the flowering plants. American Journal of Botany 82: 596-606 Reusch, T.B.H., Hukriede, W., Stam, W.T. and Olsen, J.L. 1999. Differentiating between clonal growth and limited gene flow using spatial autocorrelation of microsatellites. Heredity 83: 120-126 Reusch, T.B.H. 2000. Pollination in the marine realm: microsatellites reveal high outcrossing rates and multiple paternity in eelgrass Zostera marina. Heredity 85: 459-464 Reusch, T.B.H. 2001. Fitness costs of geitonogamous selfing in a clonal marine angiosperm (Zostera marina). Journal of Evolutionary Biology 14: 129-138 Richards, A.J. 1986. Plant Breeding Systems. George and Allen and Unwin Ltd. London Richards, E., Reichardt, M. and Rogers, S. 1994. Preparation of genomic DNA from plant tissue. Current Protocols in Molecular Biology, (eds. F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl) pp 2.3.1-2.3.3, Wiley Interscience, New York. Rogers, D.L. 2000. Genotypic diversity and clone size in old growth populations of coastal redwood. Canadian Journal of Botany 78: 1408-1419. Rudi, K., Kroken, M., Dahlberg, O.J., Deggerdal, A., Jakobsen, K.S. and Larsen, F. 1997. Rapid, universal method to isolate PCR-ready DNA using magnetic beads. Biotechniques 22: 506-511 Sackville Hamilton, N.R., Schmid, B. and Harper, J.L. 1987. Life-history concepts and the population biology of clonal organisms. Proceedings of the Royal Society of London 70 Series B 232: 35-57. Sagers, C L . 1993. Reproduction in neotropical shrubs: the occurrence and some mechanism of asexuality: Ecology 74: 615-6 Salisbury, E.J. 1942. The Reproductive Capacity of Plants. Bell, London. Schlindwein, C. and Wittmann, D. 1997. Micro-foraging routes of Bicolletes pampeana (Colletidae) and bee-induced pollen presentation in Cajophora arechavaletae (Loasaceae). Botanica Acta 110: 177-183 Schmid, B. 1986. Spatial dynamics integration within clones of grassland perennials with different growth forms. Proceedings of the Royal Society London B 228: 173-186. Schmid, B. 1990. Some ecological and evolutionary consequences of modular organization and clonal growth in plants. Evolutionary Trends in Plants 4: 25-34 Schmid, B. and Harper, J.L. 1985. Clonal growth in grassland perennials: I. Density and pattern-dependent competition between plants with different growth forms. Journal of Ecology 73: 793-808. Schulke, B. and Waser, N.M. 2001. Long-distance pollinator flights and pollen dispersal between populations of Delphinium nuttallianum. Oecologia 127: 239-245 Schuster, A., Noy-Meir, I., Heyn, C C and Dafni A. 1993. Pollination-dependent female reproductive success in a self-compatible outcrosser, Asphodelus aestivus Brot. New Phytologist 123: 165-174. Silander, J. A. 1985. Microevolution in clonal plants. Population Biology and Evolution of Clonal Organisms, (eds. J. B. C. Jackson, L. W. Buss and R. E. Cook), pp. 107- 152. Yale University Press, New Haven. Skalova, H., Pechackova, S., Herben, T., Hadincova, V., and Krahulec, F. 1997. Within population genetic differentiation in traits affecting clonal growth: Festuca rubra in a mountain grassland. Journal of Evolutionary Biology 10: 383-406 Smouse, P.E. and Peakall, R. 1999. Spatial autocorrelation analysis of individual multiallele and multilocus genetic structure. Heredity 82: 561- 573 Sneath, P.H.A and Sokal, R.R. 1973. Numerical Taxonomy: The Principles and Practice of Numerical Classification. W.H. Freeman and Company, San Francisco. Snow, A. A., Spira, T. P., Simpson, R. and Klips, R. A. 1996. The ecology of geitonogamous pollination. Floral Biology: Studies on Floral Evolution in Animal Pollinated Plants, (eds. D. G. Lloyd and S. C. H. Barrett), pp 191-216. Chapman and Hall, New York. Sokal, R. R. and Rohlf, F. J. 1995. Biometry - The Principles and Practice of Statistics in Biological Research. 3rd edition. W. H. Freeman, New York. Soule, M. E. 1980. Thresholds for survival: maintaining fitness and evolutionary potential. Conservation Biology: An Evolutionary-Ecological Perspective, (eds. M. E. Soule and B. A. Wilcox), pp. 151-169. Sinauer Associates, Sunderland, Massachusetts. Stebbins, G.L. 1950. Variation and Evolution in Plants. Columbia University Press, New York. Stehlik, I. and Holderegger, R. 2000. Spatial genetic structure and clonal diversity of 71 Anemone nemorosa in late successional deciduous woodlands of Central Europe. Journal of Ecology 88: 424-435 Steinger, T., Korner, C , and Schmid, B. 1996. Long-term persistence in a changing climate: DNA analysis suggests very old ages of clones of alpine Carex curvula. Oecologia 105: 94-99 Stewart, C. 1994. Wild/lowers of the Olympics and Cascades. Nature Education Enterprises. Port Angeles, Washington Stewart, W.N. 1983. Paleobotany and the Evolution of Plants. Cambridge University Press, New York. Stuefer, J.F. 1998. Two types of division of labour in clonal plants: benefits, costs and constraints. Perspectives in Plant Ecology. Evolution and Systematics 1: 47-60. Sutherland, W. J. and Stillman, R. A. 1988. The foraging tactics of plants. Oikos 52: 239-244 Sydes, M., and Peakall, R. 1998. Extensive clonality in the endangered shrub Haloragodendron lucasii (Haloragaceae) revealed by allozymes and RAPDs. Molecular Ecology 7: 87-93 Tamm, A., Kull, K. and Sammul, M. 2001 Classifying clonal growth forms based on vegetative mobility and ramet longevity: a whole community analysis Evolutionary Ecology 15: 383-401 Tanner, J.E., 1999. Density-dependent population dynamics in clonal organisms: a modelling approach. Journal of Animal Ecology 68: 390-399 Tanner, J. E. 2001. The influence of clonality on demography: patterns in expected longevity and survivorship. Ecology 82: 1971-1981. Thompson, H. R. 1956. Distribution of distance to nth neighbor in a population of randomly distributed individuals. Ecology 37: 391-394 Thomson, J. D. and Barrett, S. C. H. 1981. Temporal variation of gender in Aralia hispida Vent. (Araliaceae) Evolution 35: 1094-1107 Thomson, J.D. 2001. How do visitation patterns vary among pollinators in relation to floral display and floral design in a generalist pollination system. Oecologia 126: 386-394 Thomson, J.D. and Plowright, R.C. 1980. Pollen carryover, nectar rewards and pollinator behavior with special reference to Diervilla lonicera. Oecologia 46: 68-74 Tiffney, B.H and Niklas, K.J. 1985. Clonal Growth in Land Plants. Population Biology and Evolution of Clonal Organisms, (eds. J.B.C Jackson, L.W. Buss and R.E. Cook), pp 35-66. Yale University Press, New Haven. Torimaru, T., Tomaru, N., Nishimura, N. and Yamamoto, S. 2003. Clonal diversity and genetic differentiation in Ilex leucoclada M. patches in an old-growth beech forest. Molecular Ecology 12: 809-818 Turner, M.E., Stephens, J.C. and Anderson, W.W. 1982. Homozygosity and patch structure in plant populations as a result of nearest-neighbor pollination. American Naturalist 79: 203-207. Turkington, R. and Harper, J. L. 1979. The growth, distribution and neighbour 72 relationships of Trifolium repens in a permanent pasture. I. Ordination, pattern and contact. Journal of Ecology 67:201-208. Urbanska, K.M. 1989. Biology of asexually reproducing plants. Biological Approaches and Evolutionary Trends in Plants, (ed. S. Kawano), pp. 229-254. Academic Press, New York. Ushimaru, A. and Kikuzawa, K. 1999. Variation of breeding system, floral rewards, and reproductive success in clonal Calystegia spp. (Convolvulaceae). American Journal of Botany 86: 436-446 Van der Merwe, M., Winfield, M.O., Arnold, G.M. and Parker, J.S. 2000. Spatial and temporal aspects of the genetic structure of Juniperus communis populations. Molecular Ecology 9: 379-386 van Groenendael, J and de Kroon, H . 1990. Preface. Clonal growth in plants: regulation and function, (eds. J. van Groenendael and H. de Kroon, H), pp. 7-10. SPB Academic Publishing, The Hague. van Groenendael, J., Klimes, L., Klimesova, J. and Hendricks, R. 1996. Comparative Ecology of clonal plants Philosophical Transactions of the Royal Society of London 351: 1331-1339 Vange, V. 2002. Breeding system and inbreeding depression in the clonal plant species Knaulia arvensis (Dipsacaceae): implications for survival in abandoned grassland. Biological Conservation 108: 59-67 Vekemans, X. 2002. AFLP-SURV version 1.0. Distributed by the author. Laboratoire de Genetique et Ecologie Vegetale, Universite Libre de Bruxelles, Belgium. Venkataraman, G.S., Goyal, S.K., Kaushik, B.D. and Roychoudhury, P. 1974. Algae: Form and Function . Today and Tommorrow's printers and publishers, Faridabad. Verburg R., Maas, J. and H.J. During. 2000. Clonal diversity in differently-aged populations of the pseudo-annual clonal plant Circaea lutetiana L. Plant Biology 2: 646-652. Vogler, D.W. and Stephenson, A.G. 2001. The potential for mixed mating in a self-incompatible plant. International Journal of Plant Sciences 162: 801-805 Vos, P., Hogers, R., Bleeker, M., Reijans, M., van de Lee, T., Homes, M., Frijters, A., Pot, J., Peleman, J., Kuiper, M. and Zabeau, M. 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23: 4407-4414 Waser, N.M. and Price, M.V. 1991. Reproductive costs of self-pollination in Ipomopsis aggregata (Polemoniaceae). American Journal of Botany 78: 1036-1043 Watkinson, A.R. and Powell, J.C. 1993. Seedling recruitment and the maintenance of clonal diversity in plant population-a computer simulation of Ranunculus repens. Journal of Ecology 81: 707-717 Watkinson, A.R and Powell, J.C. 1997. The life history and population structure of Cycas armstrongii in monsoonal northern Australia. Oecologia 111: 341- 349 Watson, M.A. 1986. Integrated physiological units in plants. Trends in Ecology and Evolution 1: 119-123 73 Wells, G.P. and Young, A.G. 2002. Effects of seed dispersal on spatial genetic structure in populations of Rutidosis leptorrhychoides with different levels of correlated paternity. Genetical Research 79: 219-226 Wennstrom, A. and Ericson, L. 1992. Environmental heterogeneity and disease transmission within clones of Lactuca sibirica. Journal of Ecology 80: 71-77 Widen, B., Cronberg, N. and Widen, M. 1994. Genotypic diversity, Molecular markers and spatial distribution of genets in clonal plants, a literature survey. Folia Geobotanica et Phytotaxonomica 29: 245-263. Widen, B. and Widen, M. 1990. Pollen limitation and distance-dependent fecundity in females of the clonal gynodioecious herb Glechoma hederaceae (Lamiaceae). Oecologia 83: 191-196. Wilcock, C. and Neiland, R. 2002. Pollination failure in plants: why it happens and when it matters. Trends in Plant Science 7: 270-277 Wilcock, C C . and Jennings, S.B. 1999. Partner limitation and restoration of sexual reproduction in the clonal dwarf shrub Linnaea borealis L. (Caprifoliaceae). Protoplasma 208: 76-86 Wilkinson, M. 2001: PICA 4.0: Software and documentation. Department of Zoology, The Natural History Museum, London Williams, C F . 1994. Genetic consequences of seed dispersal in 3 sympatric forest herbs II: Microspatial genetic-structure within populations. Evolution 48: 1959-1972 Winfield, M.O., Arnold, G.M., Cooper, F., Le Ray, M., White, J., Karp, A. and Edwards, K.J. 1998. A study of genetic diversity in Populus nigra subsp. betulifolia in the Upper Severn area of the UK using AFLP markers. Molecular Ecology 7:3-10 Wolf, A. T., Harrison, S. P. and Hamrick, J.L. 2000. Influence of habitat patchiness on genetic diversity and spatial structure of a serpentine endemic plant. Conservation Biology 14:454-463. Worthen, W. B. and Stiles, E. W. 1986. Phenotypic and demographic variability among patches of Maianthemum canadense (Desf.) in central New Jersey, and the use of self-incompatibility for clone discrimination. Bulletin of the Torrey Botanical Club 113: 398-405. Ziegenhagen, B., Bialozyt, R., Kuhlenkamp, V., Schulze, I., Ulrich, A. and Wulf, M. 2003. Spatial patterns of maternal lineages and clones of Galium odoratum in a large ancient woodland: inferences about seedling recruitment. Journal of Ecology 91: 578-586. Zuljevic, A., Thibaut, T., Elloukal, H. and Meinesz, A. 2001. Sea slug disperses the invasive Caulerpa taxifolia. Journal of the Marine Biological Association of The United Kingdom 81: 343-344 74 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items