Thesis (MSc)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: Habitat fragmentation through the loss and modification of natural ecosystems poses a serious threat to biodiversity globally. Mechanisms and ecological implications of fragmentation have been extensively studied, yet new and meaningful insights continue to be produced. The highly diverse and ecologically complex fynbos shrubland communities that occur in the Cape Floristic Region of South Africa are amongst the most threatened by habitat fragmentation by urban, agricultural and silvicultural land uses and to the spread of invasive alien plants. Fynbos vegetation communities are fire-adapted and exploit the post-fire regeneration niche. Natural stochasticity in the fire regime means that these communities are temporally unstable; a factor that allows for the coexistence of such extreme diversity. Few studies have attempted to assess the influence of habitat fragmentation on this stochasticity, or to investigate the response of biological communities to the conditions that result. It is unknown whether this temporal instability will lead to delayed extinctions in fragmented communities, thus generating extinction debt. South Outeniqua Sandstone Fynbos occurs as a unique landscape mosaic of both insular and extensive habitats, suited to study through an island biogeography framework. Vegetation community stability was assessed through the comparison of historic and recent data sets. The responses of local vegetation and avifaunal communities to habitat fragmentation were assessed and compared on the basis of their differing motility. The unique habitat configurations also allowed for the investigation of extinction debt in the local communities and provided an opportunity to assess the influence of the surrounding matrix on species persistence and on connectivity in the artificial fragments. Data were analysed through simple regression analyses, modified Chi-squared tests and through ordination analyses. Ultimately the value of the artificial habitat fragments for the conservation of biological communities was assessed. Stability was observed in the vegetation species-area relationship for the natural islands and the mainland sites over twenty-two years. Smaller islands were found to receive fewer fires than large islands and the mainland. This consistently over-extended fire-return interval reduces the stochasticity of the local fire regime causing stable yet depauperate vegetation communities to result. Vegetation communities in the artificial fragments were found to hold area-related extinction debt, possibly due to the relatively long-term demographic turnover that typically occurs in fynbos. Avifaunal communities varied in their response to fragmentation relative to the vegetation. Birds – being motile – were found to be unaffected by isolation distance or surrounding matrix type – their response due more to changes experienced in the vegetation community than to geographic constraints. Fynbos-typical birds responded to the post-fire age of vegetation. Frugivorous birds and the matrix-habitat edge were identified as sources of alien and non-fynbos plant species that colonise the artificial fragments, potentially reducing the quality of these habitats for avifauna. Fragmented communities of South Outeniqua Sandstone Fynbos have the potential to function as biological reserves. This potential can be realised through the implementation of a fire regime that acknowledges the stochasticity required by the vegetation, the frequency required by the vegetation and the avifauna, and the practicality required by surrounding anthropogenic land-uses. / AFRIKAANSE OPSOMMING: Habitat-fragmentering, deur die verlies en verandering van natuurlike ekosisteme, hou 'n ernstige bedreiging in vir biodiversiteit wêreldwyd. Meganismes en ekologiese implikasies van die fragmentering is al breedvoerig bestudeer, maar nuwe en betekenisvolle insigte word voortaan vervaardig. Die hoogs diverse en ekologies komplekse fynbos-struikveld gemeenskappe wat in die Kaapse Floristiese Streek van Suid-Afrika voorkom, word meeste bedreig deur habitat-fragmentering deur verstedeliking, landbou en houtteeltkundige grondgebruike en die verspreiding van indringerplante. Fynbos gemeenskappe is aangepas tot vuur en die uitbuiting van post- vuur herlewing nis. Natuurlike stogastisiteit in die vuur-bedeling beteken dat hierdie gemeenskappe tydelik onstabiel word, 'n faktor wat die mede-bestaan van uiterste diversiteit moontlik maak. Min studies het al die invloed van habitat-fragmentering op hierdie stogastisiteit ondersoek, of die reaksie van biologiese-gemeenskappe tot die voortspruitende omstandighede. Dit is onbekend of die tydelike onstabiliteit sal lei tot vertraagde uitsterwing in gefragmenteerde gemeenskappe, dus genereer uitsterwings-skuld. Suid Outeniqua Sandsteen Fynbos kom as 'n unieke landskap mosaïek van beide die eilande en uitgebreide habitatte voor, geskik om deur 'n eiland biogeografie raamwerk bestudeer te word. Plant-gemeenskappe se stabiliteit is bestudeer deur die vergelyking van historiese en onlangse data stelle. Die reaksies van die plaaslike plantegroei en voëllewe gemeenskappe tot hul habitat is bestudeer en vergelyk op grond van hul verskillende beweeglikheid. Die unieke habitat konfigurasies het ook toegelaat vir die ondersoek van uitsterwings-skuld in die plaaslike gemeenskappe en 'n geleentheid gebied om die invloed van die omliggende habitat-matriks op spesies volharding en op konneksie in die kunsmatige fragmente te bepaal. Die data is ontleed deur middel van eenvoudige regressie analises, aangepasde Chi -kwadraat toetse en deur koördinerings ontledings. Ten einde die waarde van die kunsmatige habitat fragmente vir die bewaring van biologiese gemeenskappe te bepaal. Stabiliteit in die plantspesies-area verhouding vir die natuurlike eilande en die vasteland van webwerwe oor twee en twintig jaar was waargeneem. Kleiner eilande het minder brande aangeneem as groot eilande en die vasteland. Hierdie herhaalde oor-uitbrei over-extended vuur-interval het die stogastisiteit verminder van die plaaslike vuur-bedeling wat stabiele tog spesie-arme plantegroei gemeenskappe veroorsaak het. Plantegroei gemeenskappe in die kunsmatige fragmente is bevind om spesie- area -verhouding uitsterwings-skuld te bevat, moontlik as gevolg van die relatiewe lang termyn demografiese omset wat tipies voorkom in fynbos. Voëllewe gemeenskappe het gewissel in hul reaksie tot die fragmentering relatief tot die plantegroei. Voëls – weens hul beweeglikheid – blyk om nie beinvloed te word deur isolasie afstand of omliggende habitat-matriks tipe nie - hul reaksie blyk meer asgevolg van veranderinge wat ervaar word in die plantegroei gemeenskap as geografiese beperkinge. Fynbos-tipiese voëls reageer op die post- vuur ouderdom van plantegroei. Vrugte-etende voëls en die habitat-matriks rand is geïdentifiseer as bronne van uitheemse en nie-fynbos plantspesies wat die kunsmatige fragmente koloniseer, wat potensieel die kwaliteit van hierdie habitatte vir voëls verminder. Gefragmenteerde gemeenskappe van Suid Outeniqua Sandsteen Fynbos het die potensiaal om as biologiese reserwes te funksioneer. Hierdie potensiaal kan verwesenlik word deur die implementering van 'n vuur-bedeling wat erkenning verleen aan die stogastiesiteit vereis deur die plantegroei, die frekwensie wat deur die plantegroei en die voëllewe vereis word , en die praktiese vereistes van die omliggende menslike grondgebruike.

The buffy flower bat (Erophylla sezekorni) is a neotropical leaf-nosed bat (Phyllostomidae) that is endemic to the Greater Antilles. Although this species is one of the most common and abundant species of mammals in the West Indies, very little is known about its ecology and evolution. To address this deficiency, I studied the genetic structure and mating system of the buffy flower bat on several islands throughout its range, focusing a more intensive study on the island of Exuma, Bahamas. I first studied the effects of ocean barriers on genetic diversification within Erophylla and two related endemic genera of endemic West Indian bats, Brachphylla, Phyllonycteris (Chapter II). I found evidence that ocean barriers inhibit gene flow and promote speciation within these genera. Focusing on genus Erophylla (Chapter III), I found that ocean channels usually act as barriers to gene flow among island populations within species. However, relatively shallow and narrow ocean channels formed semi-permeable barriers allowing gene flow between some island populations. Within the buffy flower bat, Erophylla sezekorni (Chapter IV), genetic diversity of mitochondrial DNA fragments was positively correlated with island size, with small islands having reduced genetic diversity. However, genetic diversity at several nuclear microsatellite loci was not correlated with island area and levels of genetic diversity were high for most island populations. In addition, island populations within the Great Bahamas Bank and Little Bahamas Bank showed high levels of gene flow between islands and showed no evidence of genetic bottlenecks. Populations of E. sezekorni on Exuma (Chapters IV and V) exhibited a polygynous mating system that included vigorous visual, acoustic, and olfactory male display behaviors. However, the social structure that I observed had a negligible effect on genetic diversity and genetic structure within these populations. Overall, the buffy flower bat exhibits very few of the genetic symptoms of island life, such as reduced genetic diversity and increased genetic isolation, and is evolutionarily adapted to persist on small oceanic islands.

We report a new chameleon-like Anolis species from Hispaniola that is ecomorphologically similar to congeners found only on Cuba. Lizards from both clades possess short limbs and a short tail and utilize relatively narrow perches, leading us to recognize a novel example of ecomorphological matching among islands in the well-known Greater Antillean anole radiation. This discovery supports the hypothesis that the assembly of island faunas can be substantially deterministic and highlights the continued potential for basic discovery to reveal new insights in well-studied groups. Restricted to a threatened band of midelevation transitional forest near the border of the Dominican Republic and Haiti, this new species appears to be highly endangered.

Most of the seafloor is soft sediment, so hard substrata are isolated and island-like. In this dissertation, I explore how species distribution patterns on isolated marine hard substrata resemble terrestrial island communities, drawing on classical island biogeography theory and assembly rules, and describe how benthic invertebrate communities assemble in these island-like habitats. Higher species richness occurred on larger substrata (dropstones and shipwrecks), paralleling terrestrial island communities. However, while larger islands have greater habitat diversity and primary productivity, marine hard substrata are simpler habitats. Greater elevation in the benthic boundary layer may expose fauna to faster current, higher food supply and larval flux. Substrata located closer together had more similar communities, another pattern that resembles terrestrial islands. Dropstone fauna had a clumped distribution, indicating that larvae may disperse among substrata located close together, resulting in similar communities. In Svalbard fjords, benthic megafaunal communities were significantly different between Arctic- and Atlantic-influenced fjords. Depth and temperature had the greatest influence, with the highest diversity occurring in cold Rijpfjorden and on the north Svalbard shelf. Recruitment in Svalbard fjords was spatially and temporally variable, with lower recruitment in Rijpfjorden than in Atlantic-influenced fjords and lower recruitment at greater depth. Most of the recruits in Svalbard fjords were fast-growing, poor-competitive opportunists. On shipwrecks, communities showed two mechanisms of colonization: mobile fauna with long-dispersing planktotrophic larvae, and encrusting fauna with lecithotrophic larvae. Encrusting species reproduce asexually to cover the wreck surface, and philopatry may build up dense populations, leading to uneven communities. On terrestrial islands, non-random co-occurrence is attributed to interspecific competition, but for marine substrata, there may not be a relationship. Fauna were distributed randomly on settlement plates in Svalbard fjords, even when interspecific competition was observed. On dropstones, some morphotypes co-occurred non-randomly in the absence of overgrowth competition. Non-random co-occurrence on isolated marine hard substrata may be a result of restricted larval dispersal (for pairs co-occurring less than by chance) or epibiontism (for pairs co-occurring more often than by chance). While species distribution patterns on island-like marine hard substrata resemble terrestrial islands, the mechanisms are not necessarily the same.

The overarching objective of this dissertation was to improve our knowledge of the relationship between seascape heterogeneity and diversity of stony coral assemblages across spatial scales. Coral communities on patch reefs in three regions of the Caribbean were used as a model system to investigating this relationship because patch reef heterogeneity could be readily quantified within the seascape using remote sensing and image analysis techniques. I began with a theoretical approach, exploring the origins of observed species diversity among coral communities at increasing spatial scales. Hierarchical sampling and null models revealed that coral diversity was governed by non-random processes at local- (10s of meters) and meso- (100s of m) scales. Spatial autocorrelation and reef heterogeneity were then investigated as potential mechanistic drivers of these non-random diversity patterns. I found limited support for spatial drivers. However, beta diversity was significantly correlated to metrics of reef heterogeneity (measured as reef size, spatial configuration, and complexity), indicating that differences in reef heterogeneity were making a disproportionate contribution to the overall coral community diversity. The relationship between corals and reef heterogeneity was found to be both scale-dependent and region dependent. This theoretical approach was followed by a manipulative approach using an existing artificial patch reef array to experimentally test the influence of reef spatial configuration and topographical complexity on local diversity. Corals were most sensitive to reef size and secondarily reef configuration within the seascape. Unlike reef fishes, reef complexity did not emerge as a strong predictor of the coral community composition in either the observational data or the experimental manipulation. These observational and experimental explorations of the relationship between corals and habitat reveal that intra-habitat variability (i.e. differences between patch reefs) can influence the diversity and abundance of corals. I then focused on applying this improved theoretical understanding towards improving coral management efforts. I present a new methodology to assess the efficacy of marine reserve effects by controlling for natural seascape variation within and beyond the reserve boundary, and I quantified the bias of underestimating coral diversity by using conventional reef monitoring protocols that ignore differences in reef size. In conclusion, I demonstrate empirically that seascape attributes of reef heterogeneity can contribute to coral diversity at relatively small spatial scales (<1km) and can affect corals with different life history traits in different ways. Hence, management and conservation efforts must consider the role of these meso-scale spatial metrics to influence the structure of the coral assemblage at the local scale.

Rhadine is a Nearctic lineage of flightless ground beetles in the tribe Platynini notable for the slender and elongate habitus of the adults and, in the Southwest, the habitat preferences of many species, with several mountaintop endemics as well as microphthalmous species known from caves in central Texas. The genus is in need of a modern taxonomic revision as species identifications remain challenging, and a phylogenetic hypothesis for the overall structure of the group is needed in order to better understand the group's evolutionary history and test whether subterranean Rhadine are a monophyletic clade or not. To this end, a multigene phylogeny of Rhadine was inferred based on ~2.4-kb of aligned nucleotide sites from 3 molecular markers: cytochrome c oxidase subunit I (COI), 28S ribosomal DNA (28S), and carbamoylphosphate synthetase domain of the rudimentary gene (CAD). These gene fragments were obtained for 30 species or putative species of Rhadine as well as from members of their putative sister group, Tanystoma. Rhadine as currently circumscribed is reconstructed as paraphyletic with two species of subterranean beetles from caves in northern Mexico being resolved outside of Rhadine + Tanystoma. Rhadine sensu stricto (s. str.) is resolved with high support across analyses and is composed of two reciprocally monophyletic clades, clade I and II, the second of which is generally recovered in most analyses. Clade I includes those Rhadine with adult morphological characters defining the dissecta-, larvalis-, and subterranea-groups as well as a clade of macropthalmous subterranean perlevis-group species. Clade II, although not as robust as clade I, contains several surface-dwelling species from the western United States in the jejuna-, nivalis-, and perlevis-groups in addition to lineages of exclusively macropthalmous subterranean Rhadine. The troglobitic, cave-restricted, Rhadine classified in the subterranea-group are reconstructed in two different clades, and the clade contained within clade I also includes several species of large-eyed cavernicolous Rhadine. Those with a slender habitus (e.g., R. exilis, R. subterranea, R. austinica) evolved independently at least three times. Major biogeographic and evolutionary patterns based on these molecular results include: subterranea-group Rhadine north of the Colorado River in Texas (which all lack lateral pronotal setae) are found to comprise a monophyletic group, beetles in caves south of the Colorado River likely form another monophyletic group, and the "species pairs" of troglobitic Rhadine known to occur in the same caves that were sampled in our study are not resolved as each other's closest relatives suggesting that these caves were colonized independently by more than one lineage of Rhadine. The fine-scale attention given to populations of Rhadine isolated on mountain tops in the Madrean Sky Island region suggests that there is a great deal of genetic diversity among these lineages. In addition, these populations are resolved as reciprocally monophyletic with high support across all analyses. Haplotype networks constructed for these populations and compared with those of other described species for the same gene fragments reveal similar genetic distances between separate Sky Island Rhadine as compared to distances between described species from throughout the tree. Preliminary divergence time estimates of the Rhadine-Tanystoma lineage based on relaxed molecular clock analyses support a Miocene age for Rhadine and the Rhadine-Tanystoma lineage, with the crown ages of clade I and II being similar though not identical. All subterranean clade I Rhadine are dated to have begun diversifying within approximately the past 5 million years (Pliocene), an age that is compatible with the stratigraphy of the caves in the Balcones Escarpment. In addition, divergence estimates for the members of this clade support the climactic relict hypothesis, as they diversified during rapid temperature fluctuations during the Quaternary. However, the ages of the high altitude Sky Island Rhadine are estimated to be older than the most recent glacial maximum, suggesting that these distinct clades are considerably older than initially thought. We also performed character correlation tests using our phylogeny to test for patterns in form associated with cave habit and did not find statistical significance between subterranean habit and microphthalmy nor habit and development of the foveae of the mentum.Morphological characters that have been traditionally used to classify the genus into species groups were shown to be convergent in many cases. Despite these well-supported molecular clades, few morphological characters are consistent across all members, posing a challenge to the construction of identification tools. Nevertheless, a tentative update to the classification based on our findings is presented, and the future goals for reconstructing the phylogeny of Rhadine are discussed.

Islands have typically been considered isolated entities, patches of habitat surrounded by an entirely inhospitable marine or aquatic environment. However, there is increasing evidence that islands can be linked to the surrounding environment through the influx of subsidies, which may alter the relationship between species richness and island area. Little empirical work has been done to test these hypotheses in productive ecosystems, however. To better understand the effects of the influx of marine subsidies on island ecosystems, I assessed plant community responses to wrack biomass in an observational study on 74 small islands on the Central Coast of British Columbia. In Chapter 2, I focused on 1) how seaweed wrack subsidies affect the diversity of understory plant communities, 2) whether wrack subsidies affect the species-area relationship, and 3) whether the effect of wrack subsidies is mediated by landscape-scale habitat characteristics such as island area and shoreline slope. To assess the support for these hypotheses, I used selected from models that combined plant community data, remotely-sensed habitat characteristics, and shoreline wrack biomass. I used hierarchical models to provide further insight into the cross-scale influence of these factors on plot-scale responses. I found that wrack subsidies were associated with increased island-scale plant species richness. Although wrack subsidies did not alter the relationship between species richness and area on these islands, I found that smaller islands had higher levels of marine-derived nitrogen, indicating a greater influence of marine subsidies on the nitrogen pool of smaller islands. My results add to the weight of evidence that marine subsidies are drivers of large-scale patterns of species richness, and that the linkage between islands and the surrounding environment has implications for island communities. / Graduate

The subsidized island biogeography hypothesis proposes that nutrient subsidies, those translocated from one ecosystem to another, can indirectly influence species richness on islands by directly increasing terrestrial productivity. However, the lack of a formal statistical model makes it difficult to assess the strength of the hypothesis. I created a formal subsidized island biogeography model to determine how nutrient subsidies, in addition to area and distance from mainland, influence tree species richness. My model showed that an increase in terrestrial nitrogen abundance results in a decrease of tree species richness. Soil and plant δ 15N values were higher than expected and it is likely that nutrient subsidies from the marine environment are responsible for 15N enrichment. However, the range of observed nitrogen abundance is similar to inland coastal-zone forests, indicating that islands are similarly nitrogen deprived and may not be receiving enough nutrient subsidies to alter productivity. Tree species decline may therefore be more strongly related to the environmental conditions leading to patterns of nitrogen abundance rather than the abundance of nitrogen itself. Additionally, I proposed that bald eagles (Haliaeetus leucocephalus) are vectors of nutrient subsidies, depositing nutrient-rich guano at nest sites, which could alter soil chemistry and vegetation composition. In an exploratory study of seven nest sites, I found higher soil phosphorous at eagle nest sites relative to control sites (~ 33% higher). Phosphorous is a limiting nutrient in coastal temperate forests, additions help to alleviate chlorosis and slow growth especially when paired with nitrogen. Higher potassium concentration also occurred on eagle-inhabited islands but was not associated specifically with current nest sites, perhaps reflecting differential persistence of macronutrients in the soil. Despite expectations, soil δ 15N abundance was not statistically higher at eagle nest sites. Total soil nitrogen was also not statistically higher at eagle nest sites. There were no significant differences between vegetation composition at eagle nest sites and reference sites, but reference sites tended to be dominated by shrub species. Additionally, I proposed that bald eagles (Haliaeetus leucocephalus) are vectors of nutrient subsidies, depositing nutrient-rich guano at nest sites, which could alter soil chemistry and vegetation composition. In an exploratory study of seven nest sites, I found higher soil phosphorous at eagle nest sites relative to control sites (~ 33% higher). Phosphorous is a limiting nutrient in coastal temperate forests, additions help to alleviate chlorosis and slow growth especially when paired with nitrogen. Higher potassium concentration also occurred on eagle-inhabited islands but was not associated specifically with current nest sites, perhaps reflecting differential persistence of macronutrients in the soil. I expected to observe elevated nitrogen isotope signatures (δ 15N) given bald eagles’ position in the trophic web and the potential for volatilization of guano but soil δ 15N abundance was not statistically higher at eagle nest sites. Total soil nitrogen was also not statistically higher at eagle nest sites. There were no significant differences between vegetation composition at eagle nest sites and reference sites, but reference sites tended to be dominated by shrub species / Graduate

The Wolf Willow site (FbNp-26) is a multicomponent Precontact site located within the confines of Wanuskewin Heritage Park approximately 2 km north of the city of Saskatoon Saskatchewan. The site was excavated during 2010 and 2011 field seasons with the participation of The University of Saskatchewan’s archaeological field school and the Saskatchewan Archaeological Society’s field school. As a result of these excavations, 30 m2 were exposed and four distinct cultural levels were identified. These include the Plains Side-Notched complex, Prairie Side-Notched complex, McKean series, and Oxbow complex cultures. An analysis of artifacts, ecofacts, and features from each cultural level was undertaken in order to determine site usage for each time period. Ecological concepts are often used as heuristic devices in archeological studies. The theory of island biogeography and the study of patch dynamics are two concepts that can lend themselves to the archeological study of past human groups. Island biogeography was developed to explain speciation in insular environments. In archeological studies, the same mechanisms affecting speciation can be employed to study the development of culture. Patch dynamics can be used to hypothesize how resource availability affected the behavior of past populations. Using the aforementioned concepts, the Wanuskewin/Opimihaw Valley area can be viewed as a terrestrial island. The unique combination of resources both tangible and intangible combined to make the area a draw for Precontact populations for the past 6000 years. Wanuskewin continues to attract people from around the world as a centre of spiritual and cultural renewal, a world class tourism destination, and an educational facility.

The evolutionary ecology and population sytematics of the genus Phelsuma (Gekkonidae) were studied in the Seychelles. The distributions of all 18 Seychelles lizard species are described. These give little evidence for species turnover or the equilibrium model of island biogeography. Aspects of Phelsuma behaviour, social systems and reproduction were investigated. Phelsuma can digest pollen, which, with nectar may constitute a large part of the diet. On Praslin, P.sundbergi and P.astriata breed throughout the year, and partition their macro-habitat, micro-habitat and food resource dimensions such that P.sundbergi mainly occupies nectiferous and polliniferous palms and larger hardwood trees, whilst the smaller P.astriata is found on non-flowering palms and smaller trees. Interspecific competition is demonstrated by population density and habitat shifts between populations on islands in sympatry and allopatry. The biomass of Phelsuma geckos supported by coconut plantations in sympatric and allopatric situations tends to be relatively constant, such that there are either more small geckos or fewer large geckos. It is suggested that large body size is an adaptation to the defence of predictable and defensible food sources, such as palm flowers. Racial differention was investigated by multivariate morphometrics of shape, scalation and colouration, using appropriate techniques to negate ontogenetic effects. In the granitic islands, three phenetic aggregations of island populations were identified in the P.madagascariensis group, and two aggregations in the P.astriata group. Conventional taxonomic methods have failed to illustrate these patterns of phenetic variation, and new classification of island populations is proposed. Primary patterns of racial differentiation appear to have arisen from the sectioning of pre-existing clinal variation by rising Holocene sea levels, while secondary patterns relate to recent ecological selection pressures. P.abbotti populations on Aldabra and Assumption are sub-specifically distinct and not conspecific with granitic island forms. P.laticauda on the Farquhar group are probably not subspecifically distinct from Madagascan populations. All other coralline island populations appear to be recent introductions from the granitic islands.