No description available.

Thesis (PhD (Botany and Zoology))--Stellenbosch University, 2011. / ENGLISH ABSTRACT: With the current global decline in pollinators, and the concurrent decline in plant species, pollination research is becoming increasingly important. However, studies outside Europe and North-America and on groups other than insects are needed to make generalisations possible. In this thesis I study how pollination structures plant and bird communities in a biodiversity hotspot, the Cape Floristic Region of South Africa. I show that bird-plant pollination mutualisms are an important ecological factor structuring ornithophilous Proteaceae and nectar-feeding bird communities. This close association between plant and bird communities suggests an important role for community wide pollination mutualisms. How these mutualisms disassemble in reaction to a range of anthropogenic impacts is determined. Firstly, I use experimental manipulation of honeybee density to test whether honeybee farming affects nectar-feeding birds. Hive addition increased honeybee abundance far above natural levels but nectar-feeding bird pollinators were not consistently affected. Secondly, I document the impact of a two lane tar road on the bird pollination community. The two-fold decline found in pollination along roadsides, should have important implications for the way we view and manage road verges for ecological processes. Thirdly, I investigated how fragmentation affects bird-pollination communities by assessing an endangered, bird-pollinated plant, Brunsvigia litoralis. The only flower visitor at the urban sites, the shorter billed Greater Double-collared Sunbird is unable to access the nectar due to a long perianth tube. The longer billed Malachite Sunbird was the sole pollinator of B. litoralis at the rural site, significantly increased seed set. The lack of ecological analogs in these urban fragments might place pollinator specialist plants, such as B. litoralis, at risk. Fourthly, fire is a frequent disturbance in communities of bird-pollinated plants. In a before/after fire observation study and a burnt/unburnt transplant study, birds visited flowers in the “before fire” and “unburnt” areas only. The results are surprising given the large number of bird-pollinated plants flowering in the early post-fire vegetation. Lastly, I find that alien invasive plant species are incorporated into the native pollination community in a spectacular way; sunbirds adapt to a hummingbird-like, hovering lifestyle to obtain nectar. Alien invasive plants greatly increase nectar-feeding bird abundance; in turn, birds enhance seed set in these alien plants. I conclude by asking whether the disassembling of bird pollination communities really matters. To answer this question I report on a decade of demographic data on the geophytic bird-pollinated Brunsvigia orientalis. In the demographic analysis, the elasticity component for reproduction was more important than expected for a long lived plant. Reduced population growth in the shade and a large investment in a winged inflorescence, suggest B. orientalis is a light demanding, well dispersed, gap colonising species. The link between pollination and seed has been made before, but I take this one step further and show that pollination intensity predicts population growth rate. By linking plant demography and pollination, I was able to predict the future of plant populations under variable pollination conditions. The disassembly of bird pollination communities only becomes important for population persistence once the mutualism has almost entirely broken down. / AFRIKAANSE OPSOMMING: Met die huidige globale afname in bestuiwers en die gelyktydige afname in plant spesies, word bestuiwing navorsing toenemend belangrik. Studies buite Europa en Noord-Amerika en op groepe anders dan insekte is nodig on veralgemenings moontlik te maak. In hierdie tesis bestudeer ek hoe bestuiwing struktuur gee and plant en voël gemeenskappe in 'n biodiversiteit hotspot, die Kaapse Floristiese Ryk van Suid-Afrika. Ek wys dat voël-plant bestuiwings mutualismes 'n belangrike ekologiese faktor is in die strukturering van voël bestuifde Proteaceae gemeenskappe en nektar-etende voël gemeenskappe. Hierdie noue assosiasie tussen plant en voël gemeenskappe impliseer 'n belangrike rol vir gemeenskapwye bestuiwings meganismes. Ek bepaal hoe hierdie mutualismes aftakel in reaksie op 'n verskeidenheid van antropogeniese impakte. Eerstens gebruik ek 'n eksperimentele manipulasie van heuningby getalle om te toets of bye boerdery nektar-etende voëls affekteer. Byekorf toevoeging het heuningby getalle laat toeneem tot ver bo natuurlike vlakke maar nektar-etende voël bestuiwers is nie konsekwent beïnvloed nie. Tweedens dokumenteer ek die impakte van 'n twee baan teerpad op die voël bestuiwings gemeenskap. Die twee-malige afname in bestuiwing langs paaie sal belangrike implikasies hê vir die manier hoe ons pad reserwes sien en bestuur met betrekking tot ekologiese prosesse. Derdens bestudeer ek hoe fragmentasie die voël-plant gemeenskappe affekteer deur die bedreigde voël-bestuifde Brunsvigia litoralis te assesseer. Die enigste besoeker in die meer stedelike area, die Groot-rooibandsuikerbekkie, wat 'n korter snawel het, is nie in staat om die nektar te bereik nie, weens 'n te lang blombuis. Die Jangroentjie suikerbekkie met sy langer snawel is die enigste bestuiwer van B. litoralis in die meer landelike area, met 'n betekenisvolle vermeerdering in saad vorming. Die gebrek aan ekologies analogiese spesies in die stedelike fragmente kan 'n risiko inhou vir bestuiwer gespesialiseerde plante soos B. litoralis. Vierdens, vuur is 'n gereelde versteuring van voël-plant gemeenskappe. In 'n voor/na vuur observasie studie en 'n brand/nie-brand verplasing studie, het voëls blomme net in die “voor brand” en “nie-brand” areas besoek. Hierdie resultate is verrassend siende die groot hoeveelheid voël-bestuifde plante wat blom direk na brande. Laastens het ek gevind dat uitheemse indringer plante geïnkorporeer word in die inheemse bestuiwers gemeenskappe op 'n skouspelagtige manier; suikerbekkies pas aan tot 'n kolibri-tipe, fladderende lewenswyse om nektar te bekom. Uitheemse indringer plante het nektar-etende voël hoeveelhede laat toeneem; in reaksie het voëls saad opbrengs vermeerder. In konklusie vra ek of hierdie aftakeling van die voël bestuiwers gemeenskap belangrik is. Om hierdie vraag te antwoord assesseer ek 'n dekade van demografiese data van die geofietiese, voël-bestuifde plant, Brunsvigia orientalis. In die demografiese analises was die elastisiteit komponent van reproduksie belangriker as verwag vir 'n langlewende plant. Verminderde populasie groei in die skaduwee en 'n hoë investering in 'n gevlerkte bloeiwyse suggereer dat B. orientalis 'n lig afhanklike, goed verspreide, gaping koloniserende spesie is. Die skakel tussen bestuiwing en saadvorming is voorheen gemaak, maar ek neem dit een stap verder en wys dat bestuiwings intensiteit populasie groeikoers voorspel. Deur plant demografie en bestuiwing te koppel was ek in staat om die toekoms van populasies onder variërende bestuiwings kondisies te voorspel. Die aftakeling van voël bestuiwings gemeenskappe word slegs belangrik vir populasies se voortbestaan wanneer die mutualisme amper heeltemal verdwyn het.

Biodiversity, ecosystems, industry and human health are threatened by invasive plant species. The costs of mitigating damages run into billions of pounds per annum. Fundamental to the control of invasive plant species is an ability to predict which species will become invasive. Yet identification of predictive differences between invasive and non-invasive species has proven difficult to pinpoint. In this thesis I identify several weaknesses within published literature, and using field experiments and meta-analyses we address these to find consistent predictors of invasiveness amongst plants. Specifically, I recognize that predictors of invasiveness can be identified by studying plant species in the native range because species may undergo phenotypic and demographic changes following naturalization (Chapters 2 – 5). I also recognize the importance of comparing globally invasive and non-invasive species, and the importance of accounting for phylogenetic relationships so as not to inflate or conceal differences (Chapters 2 – 4). Finally, I investigate whether particular analyses are more appropriate for investigating life history and demographic differences (Chapter 5). This thesis comprises an introductory chapter (Chapter 1), four data chapters (Chapters 2 - 5) and a general discussion (Chapter 6). Chapters 2 and 3 compare life history traits of plant species known to be invasive elsewhere, with their exported but non-invasive sympatric relatives in the native range. Chapter 4 utilizes Population Projection Matrices held within the COMPADRE Plant Matrix Database, to compare demographic projections of stable and transient dynamics of invasive and non-invasive plants; and Chapter 5 compares ten metrics, derived from Population Projection Matrices, of seven invasive species between the native and invaded range to determine if there are demographic or life history differences that facilitate invasion, and to identify those analyses that are most likely to reveal such differences. I find reproductive capacity to be a predictor of invasiveness, and that analyses of transient dynamics are more likely than analyses of projected stable dynamics to reveal demographic or life history differences between invasive and non-invasive species or populations of plants. I discuss these findings in the context of invasive risk assessment protocols and highlight future research opportunities.

Understanding factors limiting population growth is crucial to evaluating species persistence in changing environments. I used Integral Projection Models (IPMs) to elucidate the role of biotic interactions and disturbance on population growth rate in two plants: Helianthemum squamatum, a perennial endemic to gypsum habitats in central Spain, and Liatris ohlingerae, a long-lived perennial endemic to the Lake Wales Ridge of central Florida. In H. squamatum, there was a strong positive effect of trampling in the site with the highest plant density and moderate positive effects of seed addition in the site with the lowest plant density. Differences in treatment effectiveness between sites may represent a shift from seed to microsite limitation at increasing densities. Additionally, a distinct drop in population growth rate occurred in the hottest and driest year (2009-10). In Liatris ohlingerae, roadside populations had consistently higher population growth rates than scrub populations. A modest negative effect of time-since-fire was observed in plants that did not experience herbivory. Both habitat and time-since-fire showed distinct interactions with vertebrate herbivory, with herbivory increasing the difference in growth rate between habitats and decreasing the difference between time-since-fire classes. The direct effect of herbivory was negative in all environmental combinations except in long unburned populations. These results demonstrate the importance of considering environmental interactions when constructing population models, as well as the validity of using IPMs to assess interactions in species with differing life histories.

No description available.

No description available.

Thesis (Ph. D.)--Miami University, Dept. of Botany, 2006. / Title from second page of PDF document. Includes bibliographical references.

Linking environmental variation to population dynamics is necessary to understand and predict how the environment influences species abundances and distributions. I used demographic, environmental and trait data of forest herbs to study effects of spatial variation in environmental factors on populations as well as environmental change in terms of effects of forest succession on field layer plants. The results show that abundances of field layer species during forest succession are correlated with their functional traits; species with high specific leaf area increased more in abundance. I also found that soil nutrients affect vegetative and flowering phenology of the forest herb Actaea spicata. The effect of nutrients shows that a wider range of environmental factors than usually assumed can influence plant phenology. Moreover, local environmental factors affected also the demography of A. spicata through effects on vital rates. An abiotic factor, soil potassium affecting individual growth rate, was more important for population growth rate than seed predation, the most conspicuous biotic interaction in this system. Density independent changes in soil potassium during forest succession, and to a lesser extent plant population size dependent seed predation, were predicted to alter population growth rate, and thereby the abundance, of A. spicata over time. Because these environmental factors had effects on population projections, they can potentially influence the occupancy pattern of this species along successional gradients. I conclude that including deterministic, as opposed to stochastic, environmental change in demographic models enables assessments of the effects of processes such as succession, altered land-use, and climate change on population dynamics. Models explicitly incorporating environmental factors are useful for studying population dynamics in a realistic context, and to guide management of threatened species in changing environments.