The decline of coral reefs is well documented, yet a detailed understanding of the processes involved in the establishment, persistence, and ecology of the coral-dinoflagellate associations still remains largely unknown. The advent of molecular techniques has resulted in significant advances in understanding the molecular diversity present of symbiotic dinoflagellates from the genus Symbiodinium, but information concerning the functional, ecological, and biogeographical significance of this expanding symbiont diversity remains limited. This thesis therefore used molecular methodologies to uncover Symbiodinium diversity in Stylophora pistillata, Pocillopora damicornis, and Seriatopora hystrix at ecological scales, in response to thermal stress, and to long-term environmental shifts. In addition, all the molecular methods currently used in Symbiodinium research are critically reviewed to provide an important baseline for future studies. The application of ITS2-DGGE coupled with the integration of alternate speciation concepts and analyses showed great merit in assessing Symbiodinium diversity in S. pistillata, P. damicornis, and S. hystrix along a depth gradient (3 m to 18 m) at Heron Island on the Great Barrier Reef (Australia), and emphasizes that sampling regimes should focus on the role of symbionts within their functional habitat. S. pistillata associates with symbionts C78 or C35/a in shallow areas and C79 in the deep, but also harbors a generalist type C8/a that can be found at all depths. P. damicornis harbors C42/a in shallow areas while C33/a is generally found in deeper reef zones, although it is occasionally observed in the shallows. On the other hand, S. hystrix only harbors a single symbiont (C3/t) at all depths. The data from Chapter 2 therefore shows that closely related symbionts within a single clade can diverge rapidly under influence of ecological differentiation whereby each symbiont represents a separately evolving lineage that occupies a specific ecological niche. As such, closely related symbionts are likely to have evolved specializations that optimize performance within their environmental range. Previous studies have sought to explain the bleaching susceptibility of scleractinian corals as a function of the presence or absence of six major clades of Symbiodinium. In chapter 3 it is shown that sub-cladal types of clade C in S. pistillata differ in their response to thermal stress, and these differences are as large as those previously reported between different clades. Molecular (ITS2-DGGE) data is integrated with physiological measurements (PAM fluorometry, host protein, symbiont cell density) to investigate the response to stress (bleaching) and is directly related to fine-scale differences in symbiont-types. This suggests that the cladal distinction of Symbiodinium is insufficient to explain the highly variable responses commonly seen in reef-building corals. Furthermore, the results highlight that shifts in symbiont community within a host population are due to differential mortality rather than the uptake of novel symbionts. It therefore appears that changes in the thermal tolerance of corals by acquiring novel more resistant Symbiodinium to meet the challenges of global warming may be restricted, and as such cannot be expected to prevent large-scale reef degradation. Whilst most Symbiodinium cladal studies have focused on bleaching, Chapter 4 combines molecular (ITS2-DGGE) and physiological analyses (PAM fluorometry) to assess the flexibility of the coral–symbiont assemblages in S. pistillata, P. damicornis, and S. hystrix when faced with long-term shifts in key environmental conditions. To test this, a 32 month reciprocal transplant experiment was set-up on Heron Island, whereby corals were transplanted to a new light environment, i.e. shallow to deep, or deep to shallow. Although some host-symbiont combinations were able to shuffle sub-cladal symbiont types, almost all colonies reverted back to their original type within 7 to 12 months. Interestingly, transplanted colonies showed a broad acclimatory response by adjusting their physiological responses to those of the control colonies at the transplanted depth. However, those that persisted with sub-optimal symbiont types suffered disproportionate rates of mortality at the onset of additional stress (bleaching). This suggests that, despite their large acclimatory capacity, the holobiont was likely living at the limits of its tolerance range. As such, it appears that coral species cannot readily form a novel symbiotic unit by changing their symbiotic partner in response to prolonged periods of change, and it is therefore unlikely that they will be able to adjust their symbionts in an attempt to cope with changing global conditions. Finally, Chapter 5 reviews the different DNA markers (18S, 28S, ITS1, ITS2, cp23S) and screening methods (RFLP, LICOR, SSCP, DGGE, cloning-sequencing) currently applied to uncovering Symbiodinium diversity. Current rDNA markers are used to identify ancient or more recent evolutionary lineages and separate symbionts into broad groups such as clades, as well as uncovering fine-scale differences between ecologically different sub-cladal types. The 18S and 28S rDNA in combination with RFLP are appropriate for studies that focus on large groupings at the cladal level and may uncover broad biogeographical. However, studies on the ecology or physiology of host-symbiont associations clearly benefit from the capacity to detect the full level of variability present within a community (e.g. 28S-SSCP, ITS2-DGGE). Despite the suitability of the rDNA, it is an imperative that sequence information is combined with ecological data in order to accurately predict how each hostsymbiont combinations responds as a unit that is optimized to function within the range of its distribution. Symbiont types present in individual hosts are shown to be distinct cohesive groups that are not interchangeable on an ecological, functional and evolutionary scale. In summary, this thesis expands on the current knowledge of the role of Symbiodinium on coral reefs and introduces a number of novel concepts. It is imperative that effort and resources continue to be channelled to combine genetic and ecological studies on Symbiodinium, as they appear to be an important factor driving responses of their host. Although diversity assessments would benefit from the development of a single-copy molecular marker, it is important to continue utilizing current methodologies to increase our knowledge of coral-symbiont diversity if we are to understand and manage coral reefs in the short-term as environmental conditions continue to change.

The purpose of this study was to use molecular techniques and morphological measurements to set out to find whether a hybrid zone exists between two coastal populations of the common gecko (Hoplodactylus maculatus), on the Wellington south coast. I collected geckos from five sites in a coastal transect from the population of small geckos to the large geckos. Using four genetic loci, one mitochondrial (16S) and three nuclear (Rag-1, Rag-2, C-mos), I was able to determine that the coastal populations do have geneflow, however each population maintains some unique alleles. Morphological evidence reveals a significant difference in gecko sizes from Turakirae Head and those caught at Ocean Beach, separated by just 15 km. Adult geckos at Turakirae Head are on average 10mm smaller (snout-to-vent) than adult geckos at Ocean Beach, representing almost a doubling in average weight. The centre of the steep frequency clines of four characters is coincident and the widths are concordant. The narrower morphological clines indicate stronger selection on the size of the gecko, than on genetic loci.

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / This work constitutes a detailed study of the spores, of the ferns and fern allies, both native and adventive, which grow in New Zealand, from the Kermadec Islands in the North, to the Chatham Islands in the East and the Subantarctic Islands in the South. Twenty live families with sixty five genera, two hundred and eleven species (of which c.20 are introduced) and three subspecies are included. Seven species are heterosporous the remainder are homosporous. Trilete spores are found in c.104 species representing c.30 genera. Perine is present in most taxa (with the exception of Gleichenia) and ranges from the large and sac-like form as seen in Cyathea smithii, to the thin and reduced form seen in Adiantum. Monolete spores are found in c.109 species representing c.37 genera (two genera are included twice, Isoetes which has trilete megaspores and monolete microspores and Lindsaea which has both monolete and trilete species). Perine is present in most taxa (except Sticherus spp.) and ranges from a fine deposit as seen in the Psilotaceae, to an enlarged form, heavily winged, as seen in the Aspleniaceae. Sculpture in all taxa (with the exception of the Lycopodiaceae and some members of the Ophioglossaceae, which may have ornamentation distributed distally), is present on both distal and proximal faces. One adventive taxon Equisetum arvense L. has an unusual circular laesura and elaborate elaters. Light micrographs of acetolysed and fresh spores along with scanning electron micrographs, are included for each taxon. Keys presented, are based on gross spore morphology and are applicable to fresh and acetolysed material. Descriptions include a list of synonymous species, details of spore shape, laesura/ae details, perine/exine sculpture and thickness (where sections allow), size (measured from n=50 spores per population), for samples treated with cotton blue (lactophenol aniline blue formula), mounted in glycerine jelly and acetolysed samples, mounted in silicone oil. Percentage size differences for acetolysed material mounted in glycerine jelly are also noted, along with previous spore dimensions recorded in the literature. Descriptions also contain chromosome number, where known, a list of previously recorded descriptions, a brief indication of geographic location and a list of vouchers for each sample. Experiments on the size and morphology of fern spores in reaction to different preparation techniques are discussed. Fresh spores of seven species, four trilete (Adiantum fulvum, Cyathea smithii, Hymenophyllum flabellatum and Lycopodium volubile) and three monolete (Blechnum chambersii, Paesia scaberula and Tmesipteris elongata), were assessed wth regard to the effect of three standard pretreatments (lactophenol aniline blue, 10% KOH, acetolysis) and two mounting media (glycerine jelly and silicone oil). Changes in morphology and size of the various wall layers were noted in comparison to spores observed fresh in water. Results indicate that variation includes shrinkage e.g. silicone oil and expansion of both exine and perine in glycerine jelly. Both effects are modified by previous treatments.

The role of, and response to, disturbance, in the dynamics of the mountain beech forest at western Mt Ruapehu, was examined, using palynology, dendrochronology, and vegetation survey. The pollen record indicates that, two thousand years ago, the west Ruapehu forest was dominated by Nothofagus solandri. About 1800 years B.P., the eruption of Lake Taupo devastated the forest. Though Halocarpus spp. and Phyllocladus asplenifolius were important early colonizers, Nothofagus solandri re-established itself close to its present limits after the eruption. Fire, from about 650 years ago onwards, did not affect the beech forest, but did affect the coniferous vegetation on the ring plain near the mountain. The age structure of the forest shows that there was a large disturbance event shortly before 1740. A dieback occurring about 1969 appears to have been the largest episode of mortality since 1740, larger than the dieback episode described by Cockayne at the beginning of this century. Dieback occurred as a short peak-period of Nothofagus solandri death about 1969. It was the large trees of the mature cohort which tended to be killed by dieback rather than the smaller individuals. Tree-ring analysis, showed that narrow tree-rings occurred in beech in the 1960's. Extreme rainfall years in the early 1960's may have put the beech trees under stress, making them susceptible to dieback. The drought in 1969 may have precipitated extensive mortality. The characteristics of N. solandri rings from Ruapehu are similar to ring characteristics of that species from South Island studies. The 1982 cyclone had a noticeable impact on the forest, affecting some areas severely. However, the impact was relatively minor compared to the influence of dieback. The mean density of trees >= 10cm dbh was 520 stems/ha., of which 220 stems/ha. were Nothofagus solandri. The mean basal area was 22m2 /ha., of which 11m2 /ha. was N. solandri. The basal area is very low compared to that in other N.Z. forests. N. solandri, Griselinia littoralis, Phyllocladus asplenifolius, Podocarpus hallii, and Libocedrus bidwillii, made up 35, 22, 16, 9, and 8 percent, respectively, of the total density. Basal area, density, and species composition varied as much within sites, as between them. The shrub layer formed a large part of the vegetation at west Ruapehu. Coprosma species were particularly abundant. There is sufficient regeneration to indicate that a new cohort of N. solandri is becoming established in most areas. Some areas will probably remain in shrubland for many decades. The result of dieback and windthrow has been to virtually eliminate the canopy of large old beech trees, and to increase the heterogeneity of stand structure. N. solandri seedling densities are not as high as those found in many South Island beech forests. Nothofagus solandri is growing faster than are most of the co-existing tree species, and it appears likely that it will continue as the dominant tree species. Implications for forest management are discussed.

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / This work constitutes a detailed study of the spores, of the ferns and fern allies, both native and adventive, which grow in New Zealand, from the Kermadec Islands in the North, to the Chatham Islands in the East and the Subantarctic Islands in the South. Twenty live families with sixty five genera, two hundred and eleven species (of which c.20 are introduced) and three subspecies are included. Seven species are heterosporous the remainder are homosporous. Trilete spores are found in c.104 species representing c.30 genera. Perine is present in most taxa (with the exception of Gleichenia) and ranges from the large and sac-like form as seen in Cyathea smithii, to the thin and reduced form seen in Adiantum. Monolete spores are found in c.109 species representing c.37 genera (two genera are included twice, Isoetes which has trilete megaspores and monolete microspores and Lindsaea which has both monolete and trilete species). Perine is present in most taxa (except Sticherus spp.) and ranges from a fine deposit as seen in the Psilotaceae, to an enlarged form, heavily winged, as seen in the Aspleniaceae. Sculpture in all taxa (with the exception of the Lycopodiaceae and some members of the Ophioglossaceae, which may have ornamentation distributed distally), is present on both distal and proximal faces. One adventive taxon Equisetum arvense L. has an unusual circular laesura and elaborate elaters. Light micrographs of acetolysed and fresh spores along with scanning electron micrographs, are included for each taxon. Keys presented, are based on gross spore morphology and are applicable to fresh and acetolysed material. Descriptions include a list of synonymous species, details of spore shape, laesura/ae details, perine/exine sculpture and thickness (where sections allow), size (measured from n=50 spores per population), for samples treated with cotton blue (lactophenol aniline blue formula), mounted in glycerine jelly and acetolysed samples, mounted in silicone oil. Percentage size differences for acetolysed material mounted in glycerine jelly are also noted, along with previous spore dimensions recorded in the literature. Descriptions also contain chromosome number, where known, a list of previously recorded descriptions, a brief indication of geographic location and a list of vouchers for each sample. Experiments on the size and morphology of fern spores in reaction to different preparation techniques are discussed. Fresh spores of seven species, four trilete (Adiantum fulvum, Cyathea smithii, Hymenophyllum flabellatum and Lycopodium volubile) and three monolete (Blechnum chambersii, Paesia scaberula and Tmesipteris elongata), were assessed wth regard to the effect of three standard pretreatments (lactophenol aniline blue, 10% KOH, acetolysis) and two mounting media (glycerine jelly and silicone oil). Changes in morphology and size of the various wall layers were noted in comparison to spores observed fresh in water. Results indicate that variation includes shrinkage e.g. silicone oil and expansion of both exine and perine in glycerine jelly. Both effects are modified by previous treatments.

The role of, and response to, disturbance, in the dynamics of the mountain beech forest at western Mt Ruapehu, was examined, using palynology, dendrochronology, and vegetation survey. The pollen record indicates that, two thousand years ago, the west Ruapehu forest was dominated by Nothofagus solandri. About 1800 years B.P., the eruption of Lake Taupo devastated the forest. Though Halocarpus spp. and Phyllocladus asplenifolius were important early colonizers, Nothofagus solandri re-established itself close to its present limits after the eruption. Fire, from about 650 years ago onwards, did not affect the beech forest, but did affect the coniferous vegetation on the ring plain near the mountain. The age structure of the forest shows that there was a large disturbance event shortly before 1740. A dieback occurring about 1969 appears to have been the largest episode of mortality since 1740, larger than the dieback episode described by Cockayne at the beginning of this century. Dieback occurred as a short peak-period of Nothofagus solandri death about 1969. It was the large trees of the mature cohort which tended to be killed by dieback rather than the smaller individuals. Tree-ring analysis, showed that narrow tree-rings occurred in beech in the 1960's. Extreme rainfall years in the early 1960's may have put the beech trees under stress, making them susceptible to dieback. The drought in 1969 may have precipitated extensive mortality. The characteristics of N. solandri rings from Ruapehu are similar to ring characteristics of that species from South Island studies. The 1982 cyclone had a noticeable impact on the forest, affecting some areas severely. However, the impact was relatively minor compared to the influence of dieback. The mean density of trees >= 10cm dbh was 520 stems/ha., of which 220 stems/ha. were Nothofagus solandri. The mean basal area was 22m2 /ha., of which 11m2 /ha. was N. solandri. The basal area is very low compared to that in other N.Z. forests. N. solandri, Griselinia littoralis, Phyllocladus asplenifolius, Podocarpus hallii, and Libocedrus bidwillii, made up 35, 22, 16, 9, and 8 percent, respectively, of the total density. Basal area, density, and species composition varied as much within sites, as between them. The shrub layer formed a large part of the vegetation at west Ruapehu. Coprosma species were particularly abundant. There is sufficient regeneration to indicate that a new cohort of N. solandri is becoming established in most areas. Some areas will probably remain in shrubland for many decades. The result of dieback and windthrow has been to virtually eliminate the canopy of large old beech trees, and to increase the heterogeneity of stand structure. N. solandri seedling densities are not as high as those found in many South Island beech forests. Nothofagus solandri is growing faster than are most of the co-existing tree species, and it appears likely that it will continue as the dominant tree species. Implications for forest management are discussed.

The decline of coral reefs is well documented, yet a detailed understanding of the processes involved in the establishment, persistence, and ecology of the coral-dinoflagellate associations still remains largely unknown. The advent of molecular techniques has resulted in significant advances in understanding the molecular diversity present of symbiotic dinoflagellates from the genus Symbiodinium, but information concerning the functional, ecological, and biogeographical significance of this expanding symbiont diversity remains limited. This thesis therefore used molecular methodologies to uncover Symbiodinium diversity in Stylophora pistillata, Pocillopora damicornis, and Seriatopora hystrix at ecological scales, in response to thermal stress, and to long-term environmental shifts. In addition, all the molecular methods currently used in Symbiodinium research are critically reviewed to provide an important baseline for future studies. The application of ITS2-DGGE coupled with the integration of alternate speciation concepts and analyses showed great merit in assessing Symbiodinium diversity in S. pistillata, P. damicornis, and S. hystrix along a depth gradient (3 m to 18 m) at Heron Island on the Great Barrier Reef (Australia), and emphasizes that sampling regimes should focus on the role of symbionts within their functional habitat. S. pistillata associates with symbionts C78 or C35/a in shallow areas and C79 in the deep, but also harbors a generalist type C8/a that can be found at all depths. P. damicornis harbors C42/a in shallow areas while C33/a is generally found in deeper reef zones, although it is occasionally observed in the shallows. On the other hand, S. hystrix only harbors a single symbiont (C3/t) at all depths. The data from Chapter 2 therefore shows that closely related symbionts within a single clade can diverge rapidly under influence of ecological differentiation whereby each symbiont represents a separately evolving lineage that occupies a specific ecological niche. As such, closely related symbionts are likely to have evolved specializations that optimize performance within their environmental range. Previous studies have sought to explain the bleaching susceptibility of scleractinian corals as a function of the presence or absence of six major clades of Symbiodinium. In chapter 3 it is shown that sub-cladal types of clade C in S. pistillata differ in their response to thermal stress, and these differences are as large as those previously reported between different clades. Molecular (ITS2-DGGE) data is integrated with physiological measurements (PAM fluorometry, host protein, symbiont cell density) to investigate the response to stress (bleaching) and is directly related to fine-scale differences in symbiont-types. This suggests that the cladal distinction of Symbiodinium is insufficient to explain the highly variable responses commonly seen in reef-building corals. Furthermore, the results highlight that shifts in symbiont community within a host population are due to differential mortality rather than the uptake of novel symbionts. It therefore appears that changes in the thermal tolerance of corals by acquiring novel more resistant Symbiodinium to meet the challenges of global warming may be restricted, and as such cannot be expected to prevent large-scale reef degradation. Whilst most Symbiodinium cladal studies have focused on bleaching, Chapter 4 combines molecular (ITS2-DGGE) and physiological analyses (PAM fluorometry) to assess the flexibility of the coral–symbiont assemblages in S. pistillata, P. damicornis, and S. hystrix when faced with long-term shifts in key environmental conditions. To test this, a 32 month reciprocal transplant experiment was set-up on Heron Island, whereby corals were transplanted to a new light environment, i.e. shallow to deep, or deep to shallow. Although some host-symbiont combinations were able to shuffle sub-cladal symbiont types, almost all colonies reverted back to their original type within 7 to 12 months. Interestingly, transplanted colonies showed a broad acclimatory response by adjusting their physiological responses to those of the control colonies at the transplanted depth. However, those that persisted with sub-optimal symbiont types suffered disproportionate rates of mortality at the onset of additional stress (bleaching). This suggests that, despite their large acclimatory capacity, the holobiont was likely living at the limits of its tolerance range. As such, it appears that coral species cannot readily form a novel symbiotic unit by changing their symbiotic partner in response to prolonged periods of change, and it is therefore unlikely that they will be able to adjust their symbionts in an attempt to cope with changing global conditions. Finally, Chapter 5 reviews the different DNA markers (18S, 28S, ITS1, ITS2, cp23S) and screening methods (RFLP, LICOR, SSCP, DGGE, cloning-sequencing) currently applied to uncovering Symbiodinium diversity. Current rDNA markers are used to identify ancient or more recent evolutionary lineages and separate symbionts into broad groups such as clades, as well as uncovering fine-scale differences between ecologically different sub-cladal types. The 18S and 28S rDNA in combination with RFLP are appropriate for studies that focus on large groupings at the cladal level and may uncover broad biogeographical. However, studies on the ecology or physiology of host-symbiont associations clearly benefit from the capacity to detect the full level of variability present within a community (e.g. 28S-SSCP, ITS2-DGGE). Despite the suitability of the rDNA, it is an imperative that sequence information is combined with ecological data in order to accurately predict how each hostsymbiont combinations responds as a unit that is optimized to function within the range of its distribution. Symbiont types present in individual hosts are shown to be distinct cohesive groups that are not interchangeable on an ecological, functional and evolutionary scale. In summary, this thesis expands on the current knowledge of the role of Symbiodinium on coral reefs and introduces a number of novel concepts. It is imperative that effort and resources continue to be channelled to combine genetic and ecological studies on Symbiodinium, as they appear to be an important factor driving responses of their host. Although diversity assessments would benefit from the development of a single-copy molecular marker, it is important to continue utilizing current methodologies to increase our knowledge of coral-symbiont diversity if we are to understand and manage coral reefs in the short-term as environmental conditions continue to change.

Tabanidae (Diptera: Brachycera), a maior família da infraordem Tabanomorpha, possui aproximadamente 4500 espécies válidas, que se distribuem por todas as regiões biogeográficas, com exceção da Antártica. São insetos cujo tamanho varia de quatro a trinta mm. As fêmas são hematófagas, e os machos visitam flores. A classificação de Tabanidae constitui um problema histórico, e apesar de um número grande de autores ter abordado o problema, poucos são os trabalhos que trataram do grupo dentro de uma perspectiva filogenética. Nesse contexto, este trabalho, junto com Lessard et al. (2013 Molec.Phylog.Evol. 68: 516540), são os primeiros tratados envolvendo a classificação da Tribo Scionini (Tabanidae: Pangoniinae) utilizando análises filogenéticas. Para o presente estudo, foram levantados caracteres da cabeça, asa e terminálias feminina e masculina dos espécimes analisados e esses caracteres foram compilados em uma matriz de 37 táxons por 44 caracteres. A matriz foi analisada com o auxílio do programa TNT. Desta análise foram obtidos quatro cladogramas igualmente parcimoniosos, sobre os quais foi utilizado o consenso estrito. No cladograma de consenso, duas subfamílias de Tabanidae são mostradas como grupos naturais (Tabaninae e Chrysopsinae), enquanto a monofilia de Pangoniinae permaneceu não resolvida na análise com pesos iguais. Quando utilizada pesagem implícita, todas as três famílias são mostradas como monofiléticas. Os Scionini se mostraram um grupo merofilético, assim como dois de seus gêneros, Scaptia e Fidena. A partir dos resultados aqui obtidos, é indicada a necessidade para a reformulação futura da delimitação dos Scionini, bem como é apontada a necessidade de revisão de alguns gêneros que compõem a subfamília Pangoniinae. / Tabanidae (Diptera: Brachycera), the most speciose family within the infraorder Tabanomorpha, comprises approximately 4500 valid species, distributed through all the biogeographical regions, except Antarctica. These insects range in length from four to tirty millimeters. Females are hematophagus and males visit flowers. The classification of Tabanidae constitutes an historical problem and, despite the fact that a great number of authors had addressed the problem, few are the works that treated the group from a phylogenetic perspective. In this context, this work, together with Lessard et al. (2013 Molec.Phylog.Evol. 68:526-540), are the first treatises dealing with the classification of the tribe Scionini (Tabanidae: Pangoniinae) based on a phylogenetic analysis. In the present work, characters from head, wing and both female and male terminalia were coded into a matrix of 37 taxa by 44 characters. The matrix was analyzed with the software TNT. From this analysis four equally parsimonious cladograms were obtained, over which the strict consensus was calculated. In the consensus cladogram, two subfamilies of Tabanidae are shown as natural groups (Tabaninae and Chrysopsinae), whereas the monophyly of Pangoniinae was not recovered with equal weights parsimony. When implied weighting was used however, all the three subfamilies are shown to be monophyletic groups. The Scionini were shown as merophyletic, as did two of its genera, Scaptia and Fidena. From the results herein obtained, it is indicated the need for the re-delimitation of Scionini. It is also argued that revisions of some genera comprised within Pangoniinae are urgent for achievement of improved taxonomic stability in this subfamily.

This thesis centres around two genera of large peltospirid gastropods (Mollusca: Neomphalina: Peltospiridae) endemic to hydrothermal vent ecosystems. One is the 'scaly-foot gastropod', an emblematic species of the Indian Ocean vents with unique dermal sclerites covering the foot like roof tiles. The other was recently discovered from expeditions to the Southern and Indian oceans, lacks sclerites and possesses large opercula. As both genera and their assigned species remained undescribed, they were formally described herein which forms a basis to understanding their biology. The 'scaly-foot gastropod' from both the Central Indian Ridge (CIR) and the Southwest Indian Ridge (SWIR) were confirmed to represent a single species and is formally named as Chrysomallon squamiferum. Through molecular genetic analyses using the COI gene, genetic differentiation between SWIR and CIR populations was detected for the 'scaly-foot gastropod'. In contrast, the peltospirids with large opercula from the East Scotia Ridge (ESR) and the SWIR proved to be two distinct species within an undescribed genus. The ESR species was formally described as Gigantopelta chessoia and the SWIR species as G. aegis. The molecular genetic analyses of the COI gene, confirmed the genetic isolation of the two and consolidated their status as separate species. A 3D tomographic model of Chrysomallon squamiferum was generated to characterise the soft anatomy and morphology as well as to understand its internal anatomy and adaptation which remained little-studied. Further to the enlarged esophageal gland already known to house chemosynthetic endosymbionts, C. squamiferum was discovered to have a hypertrophied circulatory system with a gigantic, muscular heart and large ctenidium to adapt to life in a hypoxic environment and to supply the endosymbionts with necessary chemicals. Histological examinations of the sclerites and operculum showed that it was unlikely that the sclerites originated from operculum duplication. Comparisons with polyplacophoran scales revealed starkly different secretion mechanisms despite the superficial similarity, which has implications on the placement of sclerite-bearing Cambrian taxa. Overall, the results from this thesis ascertained the systematic positions of these large-sized, enigmatic peltospirids, and led to improved understanding of their ecology and evolution. The important role of larval dispersal in maintaining metapopulations across the distribution of a vent-endemic taxa is highlighted. The adaptations of vent-endemic taxa remains little-known even in well-studied species, warranting future studies on these and other species.

Ranunculus is the largest genus in the Ranunculaceae family and comprises c. 600 species. Its distribution is almost worldwide and the largest number of species occurs in temperate zones of North and South America, Europe, Asia, Australia, New Zealand, and in the alpine regions of New Guinea. In New Zealand the genus Ranunculus contains about 41 species and is found both in lowland and alpine environments. This thesis reports a phylogentic analysis of lowland and alpine New Zealand Ranunculus, an assessment of morphological variation and species boundaries among complex alpine species and examines evidence suggesting adaptive radiation of the alpine Ranunculus lineage. Phylogenetic analysis suggests that New Zealand species of Ranunculus are not a monophyletic group. For some New Zealand species the closest affinities inferred from the analysis of nrDNA and cpDNA sequences are to species from other land masses such as Australia, the Northern Hemisphere, southern South America and islands in the southern Oceans. Contrary to Fisher’s hypothesis (1965), the Andean South American Ranunculus in the section Trollianthoideae are not closely related to the New Zealand alpine group. The Trollianthoideae section was not monophyletic and the Peruvian-Ecuadorian species in it form a lineage sister to European alpine species. Instead, aquatic and sub-aquatic species from the Euro-Mediterranean region and southern South America and the Kerguelen Island were inferred as the closest relatives to the New Zealand alpine Ranunculus; albeit this relationship was weakly supported. Findings from this study suggest that colonisation of Ranunculus into the Southern Hemisphere has been a dynamic process and several long distance dispersal events and different colonisation routes have been used. Dispersal from New Zealand to Australia and vice versa, has also been inferred. Bird transportation and oceanic currents are speculated as being the most likely vectors for long dispersal for this group. Morphological variability at the species level is a feature of several species of Ranunculus worldwide. In New Zealand, the alpine species R. insignis and R. enysii are characterised by extensive morphological variability across their distribution range. Currently, these two species include a number of geographically restricted forms that in earlier taxonomic treatments were considered as separate species. Analysis of qualitative and quantitative morphological characters using parametric and non-parametric statistical tests and multivariate analysis, habitat characterisation using environmental variables from the GIS database LENZ and molecular analyses of nrDNA and cpDNA sequences have provided a framework for interpreting and understanding the nature of this phenotypic variation. An argument based on morphological, genetic and ecological support for the reinstatement of the species R. insignis, R. lobulatus and R. monroi is presented here. The last two species may correspond to lineages of recent origin. Hybridisation and introgression between R. insignis and R. lobulatus are suggested as being responsible for intermediate phenotypes found in areas where their distribution overlaps. Morphological variability in R. enysii is inferred to have had a complex origin. The species has a disjunct distribution and events of hybridisation and/or introgression with R. monroi and R. gracilipes seems to have occurred in some of the northern and southern populations, respectively. These hybrid lineages may have swamped out pure lineages of R. enysii and eliminated the ancestral phenotype. Studies including assessment of gene flow using microsatellites, phenotypic stability under common garden condition and pollination experiments will be necessary to further test these hypotheses. Contrary to the latter two species, R. lyallii is morphologically uniform across its distribution range but genetically diverse (11 haplotypes, one of them shared with R. buchananii). Morphological stability in this species is probably explained by morphological stasis and habitat specialisation. The alpine Ranunculus group is outstanding in the New Zealand flora in terms of its great phenotypic and ecological diversity of its members. These two features plus the monophyletic nature of the group and its recent origin have suggested to previous researchers that the radiation of this group has been adaptive. Phylogenetic analysis of 20 taxa in this group using nrDNA and cpDNA sequences has shown that the group includes four lineages and that genetic diversity between the species forming each linage is low. This confirms findings from earlier studies by Lockhart et al. (2001). Cluster Analysis, multidimensional scaling analysis and histological and scanning microscopy observations of morphological and anatomical vegetative and reproductive characters were used to quantify the extent of morphological diversity in the group. Habitat diversity of this group was characterised using 16 environmental variables available from the GIS database LENZ and analysed using Canonical variates analysis. Although four habitat types were identified, there was no correlation between habitat and phenotype as predicted for an adaptive radiation. A number of alternative explanations for this lack of correspondence are discussed. The conclusion drawn from this study was that available data layers and resolution of LENZ limit the use of GIS databases for testing hypotheses of adaptation in the New Zealand Alps.