New Zealand is a land with a unique and very diversified flora which, along with the great range of habitats to be found throughout the country, has given rise to some remarkable plant communities. As a result of a considerable volume of ecological research, mainly by the efforts of Cockayne, our plant associations are well known and delimited. There have, however, been few detailed analyses of their structure and of the factors operating in any of them, Cockayne's work being primarily descriptive. From the developmental point of view the general life-history of the Kauri forest has been known for many years, but nothing more detailed is known about the changes occuring. It is for the above reason that an attempt has been made in the present research to try and elucidate some of these factors while there are still considerable areas of Kauri forest in a more or less untouched state.

Annual reproductive cycles and the related changes in the pyloric caeca of the New Zealand starfish Stichaster australis and Coscinasterias calamaria (Echinodermata, Asteroidea) were determined at Maori Bay on the west coast of Auckland. Both species were shown to have clearly defined reproductive and pyloric caeca cycles, closely repeated from year to year. Pyloric caeca cycles showed an approximate although slightly displaced inverse relationship to gonad cycles. Changes in the gonad shown by histological sections confirmed these cycles. In addition to sexual reproduction C. calamaria may reproduce asexually by autotomy. Frequency of asexual reproduction did not vary seasonally and may be related to physiological stress or mechanical damage. Oocytes of S. australis and C. calamaria were fertilized in the laboratory and the resulting larvae reared through to metamorphosed starfish. The methods used in larval culture and embryological and larval development are described. Larvae of S. australis and C. calamaria closely resemble the larvae of Asterias rubens, described by Gemmill (1914), and of other planktotrophic asteroid larvae. Abnormal larvae found in some cultures are thought to result from, unsuitable culture conditions. It was found that oocytes of both species could be cross fertilized with sperm of the other. The settling behaviour was described and substrate preferences of advanced brachiolaria larvae of S. australis and C. calamaria were determined with larvae reared in the laboratory. These results were correlated with field observations of habitat preferences of juvenile starfish. C. calamaria larvae settled on almost any hard substratum provided it was coated with a primary film. Recently metamorphosed C. calamaria could not be located at Maori Bay and it is inferred that there is low recruitment from the plankton to the Maori Bay population. S. australis larvae would only settle, metamorphose and feed on the encrusting coralline alga Mesophyllum insigne. Nursery areas of offshore boulders covered with this alga were located at Maori Bay and these were found to support populations of juvenile starfish. Mesophyllum insigne appears to provide a stable and abundant food source for juvenile S. australis. The structure of the brachiolar arms and adhesive disc of S. australis and C. calamaria was determined from light microscopy and from scanning and transmission electron microscopy. The structure of these organs was very similar in S. australis and C. calamaria. The brachiolar arms are comprised of a stem region terminating in a crown of adhesive papillae. Adhesive papillae are made up of a variety of secretory cell types. Principal among these are elongated cells producing very electron dense secretory particles. These are released at the free cell surface attached to cilia. secretory particles appear to be important in temporary attachment of the brachiolar arms to the substratum. Ciliary sense cells, possibly used in the recognition of specific substrates are located at the tip of adhesive papillae. The adhesive disc is comprised of large cells packed with secretory droplets and elongated intracellular fibres. In the attached adhesive disc secretory droplets have been lost, having formed the cement attaching the disc to the substratum. The intracellular fibres are the principal anchoring structures running from the disc surface microvilli (locked into the attachment cement) to the underlying connective tissue of the attachment stalk. S. australis juveniles reared in the laboratory were maintained on Mesophyllum insigne substrata and the growth rates and feeding behaviour of individual starfish was determined. Growth of yearly settlement cohorts on a nursery site on the shore used in conjunction with laboratory results gave a fairly clear picture of growth after settlement. Growth rates followed a typical logistic curve, growth being slow initially and becoming more rapid later. The numbers of juveniles recruited to nursery areas vary from year to year but mortality following settlement appears to be low. It was found that juvenile S. australis predate M. insigne exclusively until they reach approximately 0.8cm in diameter (7-8 months old) when they may occasionally predate juvenile perna canaliculus. The incidence of carnivorous feeding gradually increases until juveniles are approximately 2.0-2.5cm in diameter (15-18 months old) by which time they are exclusively carnivorous on small P. canaliculus. As growth continues juvenile starfish gradually migrate from nursery areas to adjacent reefs where dense beds of adult P. canaliculus occur. Starfish of this species become sexually mature when they reach approximately 8.0cm in diameter. Juvenile C. calamaria reared in the laboratory feed on the primary film coating the substratum. Field observations at Leigh showed juveniles of 0.8-0.9cm diameter to be feeding on small barnacles on the undersides of boulders. Juveniles could not be reared to this size in the laboratory and the point at which juveniles change from a herbivorous or microphagous to a carnivorous diet is still unknown. Because of frequent autotomy it was not possible to determine growth rates of C. calamaria in the laboratory. Starfish of this species become sexually mature they reach 5.0-10.0cm in diameter. / Whole document restricted, but available by request, use the feedback form to request access.

Anthocyanin concentration is an important determinant of the colour of many fruits. In apple (Malus x domestica), centuries of breeding have produced numerous varieties in which levels of anthocyanin pigment vary widely and change differently in response to environmental and developmental stimuli. The apple fruit cortex is usually colourless, although germplasm does exist where the cortex is highly pigmented due to the accumulation of anthocyanins, carotenoids or chlorophylls. From studies in a diverse array of plants species, it is apparent that anthocyanin biosynthesis is controlled at the level of transcription. In this thesis the transcript levels of the anthocyanin biosynthetic genes in a red-fleshed apple are compared with a white-fleshed cultivar. A MYB transcription factor (TF), MdMYB10, was isolated and is shown to be similar in sequence to known anthocyanin regulators in other species. Further, this TF is shown to induce anthocyanin accumulation in both heterologous and homologous systems, generating pigmented patches in transient assays in tobacco leaves and highly pigmented apple plants following stable transformation with constitutively expressed MdMYB10. Efficient induction of anthocyanin biosynthesis in transient assays by MdMYB10 was dependent on the co-expression of two bHLH proteins from apple, MdbHLH3 and MdbHLH33. A strong correlation between expression of MdMYB10 and apple anthocyanin levels during fruit development suggests that this TF is responsible for controlling anthocyanin biosynthesis in apple fruit; in the red-fleshed cultivar and in the skin of other varieties, there is an induction of MdMYB10 expression concurrent with colour formation during development. Mutations in the genes encoding for either the biosynthetic or transcriptional regulation of the anthocyanin pathway have been linked to colour phenotypes. Generally this is a loss of function resulting in a reduction or a change in the distribution of anthocyanin. The upstream regulatory region of MdMYB10 was investigated and found to contain a rearrangement. This modification is responsible for increasing the level of anthocyanin throughout the plant to produce a striking phenotype that includes red foliage and red fruit flesh. It consists of a series of multiple repeats, forming a minisatellite-like structure that comprises five direct tandem repeats of a 23 base pair sequence. This MdMYB10 rearrangement is present in all the red foliage apple varieties and species tested, but in none of the white fleshed varieties. Transient assays demonstrated that the 23 bp sequence motif is a target of the MdMYB10 protein itself, and the number of repeat units correlates with an increase in transactivation by MdMYB10 protein. The repeat motif is capable of binding MdMYB10 protein in electrophoretic mobility shift assays. Taken together, these results indicate that a rearrangement in the promoter of MdMYB10 has generated an autoregulatory locus and this autoregulation is sufficient to account for the increase in MdMYB10 transcript levels and subsequent ectopic accumulation of anthocyanins throughout the plant. Characterisation of MdMYB10 and the rearrangement in the promoter region has implications for the development of new varieties through classical breeding or a biotechnological approach. Understanding whether this mutation is simply an allele of other recently published apple MYB TFs, or if this is the only R2R3 MYB TF involved in apple anthocyanin response, is a challenge for future research.

The cell cycle is one of the fundamental processes in nature, and is primarily concerned with the faithful replication of cellular contents, followed by even division to produce two identical daughter cells. It is made up of five discrete biochemical steps, comprising the interphase (G1, S and G2) and the mitotic phase (mitosis and cytokinesis), with two major regulatory checkpoints at G1 and G2. The focus of this research is the G2 checkpoint, which ensures the successful achievement of DNA replication, prior to the initiation of mitosis. Arrest or progression is principally mediated by the CDK1/cyclin B1 complex; phosphorylation of CDK1 by wee1 kinase prevents progression to mitosis, and subsequent dephosphorylation by the CDC25 phosphatases, initially the B isoform, leads to mitotic onset. The aim of this research was the biophysical and/or biochemical characterisation of the molecular complexes that form at the G2 checkpoint to regulate entry into mitosis. CDK1 and cyclin B1 were separately expressed and purified from baculovirus-infected Sf9 cells. The wee1/14-3-3β complex was also expressed and purified, incorporating either full length wee1 or a truncated version from which the N-terminal domain of wee1 was deleted. Both exhibited wee1 kinase activity, at equivalent levels, p<0.001, with a 2.4 fold increase in kinase activity when wee1 is bound by 14-3-3β, p>0.001. Tryptic digestion of the complex indicated that its architecture was likely to be flexible and open, particularly within the N-terminal domain of wee1. CD analyses indicated that the wee1/14-3-3β complex was folded, with 30-40% α-helical content and 10-20% β-sheet content. Dissociation experiments were unsuccessful, however, indicating a high strength of interaction between wee1 and 14-3-3β. The empirical stoichiometry of the complex was determined as 1:1; subsequent native molecular weight determination suggested that the minimal functional unit is likely to be a 2:2 wee1/14-3-3β arrangement. It was proposed that the structural architecture of this complex may be similar to the serotonin N-acetyltransferase/14-3-3ζ complex. Experiments to determine the structure experimentally, using either TEM or x-ray crystallography, were unsuccessful, as the complex appeared to exhibit a high degree of flexibility in solution. CDC25B was also expressed and purified, and was found to co-purify with a putative Sf9 14-3-3 protein. Consequently, it was re-cloned to co-express with 14-3-3β, and subsequent analysis of the resulting CDC25B/14-3-3β complex indicated that the empirical stoichiometry was 1:1, with the functional organization likely to be a 2:2 arrangement. It was proposed that the structural arrangement of this complex is most likely to be similar to that of the wee1/14-3-3β complex.

The Podocarpaceae is an essentially southern hemisphere family of conifers which includes seven genera: viz. Podocarpus, Dacrydium, Phyllocladus, Microstrobos, Microcachrys, Acmopyle, and Saxegothaea. The first of these is represented in every major southern continent except Antarctica, includes some 90 species, and is sub-divided into eight sections (Buchholz & Gray, 1948). Dacrydium and Phyllocladus are both fairly widely distributed around the southern borders of the Pacific Basin and include 20 and six species respectively. All three of these genera are represented in New Zealand. The remaining four genera are all of very limited distribution, again around the borders of the South Pacific, and include only one or two species each. None is represented in New Zealand.

Natural vanilla (Vanilla planifolia, V. tahitensis) is economically important for a number of producing countries, but viral diseases are prejudicial to its successful cultivation. Techniques are required to detect viruses in vanilla plants in order to establish quarantine procedures and sources of virus-free planting material. This study contributed to the progress in managing viral diseases of vanilla by firstly identifying Cucumber mosaic virus (CMV, cucumovirus) as causing severe distortion, stunting, sterility and sometimes death of vanilla plants. Vanilla CMV isolates from French Polynesia (Pacific Ocean) and Réunion Island (Indian Ocean) were classified in CMV subgroup IB, adding to the short list of subgroup IB isolates detected outside of Asia. Two isolates were putatively classified in subgroup IA, suggesting that CMV in vanilla is still evolving and/or that vanilla is infected from different sources. Subgroup II CMV from New Zealand was also experimentally infectious to V. planifolia, showing that vanilla crops should be protected from all potential sources of CMV inoculum. Existing serological and molecular detection rests were performant for the detection of CMV directly from vanilla tissue. Secondly, this study provided the first coat protein sequence information for Vanilla mosaic virus (VanMV, Potyvirus). A Cook Islands isolate (VanMV-CI) and a French Polynesia isolate (VanMV-FP) had distinctly different coat proteins. The VanMV-FP CP N-terminus contained a stretch of amino-acid repeats (GTN) typical of natively unfolded proteins. This GTN stretch was located downstream of a DVG motif (which replaced the more common aphid transmission DAG motif), suggesting a role in improving aphid transmission, or regulating formation of the HC-virus complex. CP core nucleotide sequence identities indicated VanMV-CI and VanMV-FP were strains of Dasheen mosaic virus (DsMV). In contrast, CP amino-acid sequence homologies between VanMV- CI and DsMV were intermediate between strains and species, and CP amino-acid homologies between VanMV-FP and DsMV were typical of distinct species. In addition, VanMv-CI and VanMv-FP had characteristic 3'NTR sequences and NIb/CP cleavage sites, and only infected vanilla. Hence, it is proposed that VanMV-CI and VanMV-FP are considered new Potyvirus species and named Vanilla mosaic Cook Islands virus and Vanilla mosaic French Polynesia virus. Alternatively, the two isolates may be grouped under the name Dasheen mosaic virus-Vanilla (DsMV-V) and distinguished from Dasheen mosaic virus-Dasheen(DsMV-D). Primers to VanMV-CI and VanMV-FP were designed and permitted RT-PCR detection of the viruses directly from vanilla tissue. VanMV-Cl and VanMV-FP could be differentiated from DsMV and Watermelon mosaic virus (WMV-Tonga), and differentiated from each other by comparison of amplicon size. Long-term specific potyvirus diagnosis is however expected to be difficult due to potyviral variability in vanilla. Future research should concentrate on techniques such as microarrays to permit simultaneous detection combined with specific identification of potyvirus species. Such techniques would be beneficial to viral disease management in vanilla and many other crops.

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / Four topics concerning the cell walls of commelinid monocotyledons were investigated. First, the distribution of (1→3),(1→4)-β-glucans and (1→3)-β-glucans in the walls of different cell types in vegetative organs of barley (Hordeum vulgare) was determined using immunogold labelling with monoclonal antibodies. All walls were labelled with the antibody to (1→3),(1→4)-β-glucans, except those of the outer root cap. Two types of labelling distribution were found over primary walls: only adjacent to the plasma membrane, and throughout the wall and middle lamella. Small amounts of labelling occurred with the (1→3)-β-glucan antibody, mostly over plasmodesmata. Both antibodies labelled cell plates. Second, changes in location and density of (1→3),(1→4)-β-glucans in specific walls of barley coleoptiles during development were investigated by immunogold labelling and analysed quantitatively. In non-lignified walls from 2.5- to 5-days after germination, three different patterns of change were recognized. In most walls, there was an increasing amount of label located evenly throughout the walls (Type A pattern). Some walls showed a decrease in labelling density in the area beneath the cuticle but an increase in the area adjacent to the plasma membrane (Type B), whereas others showed decreasing amounts of label located evenly throughout the walls (Type C). In lignified walls (Type D) the label was confined mainly to the compound middle lamella. Third, the location of (1→3),(1→4)-ß-glucans in walls of vegetative organs from 9 families of the Poales (s.l.) was determined by immunogold labelling. Three types of wall-labelling pattern were identified depending on the labelling of specific walls. Type 1 (Poaceae and Flagellariaceae) had the heaviest labelling of non-lignified walls and Type 2 (Restionaceae and Xyridaceae) the heaviest labelling of lignified walls. Type 3 (Cyperaceae and Juncaceae) had only very light labelling of all walls. Walls of the Typhaceae, Sparganiaceae, and Bromeliaceae were unlabelled. Fourth, the ferulic and p-coumaric acids ester-linked to non-lignified primary walls from 11 families of commelinid monocotyledons were quantified. These acids were present in the walls of all species, with trans-ferulic acid the most abundant. ‘Driselase’ digestion and mild acid hydrolysis of isolated walls released ferulolyl oligosaccharides with structures that indicated the ferulic acid was ester-linked to glucuronoarabinoxylans.

Two amylin responsive proteins, here designated ARP1 and ARP2, were discovered from rat skeletal muscle through two dimensional gel electrophoresis analysis. ARP1 was detected only in amylin-stimulated muscles where the insulin-stimulated glucose incorporation into glycogen was inhibited. This protein incorporated 32Pi but not [35S]-methionine in the metabolic labeling experiments. Subsequent molecular characterisation revealed that ARP1 was a novel monomeric form (designated form 1) of protein p20, and two other monomeric forms (designated forms 2 and 3 respectively) of protein p20 were also characterised. The production of ARP1 was not affected by the presence of insulin, but calcitonin gene-related peptide (CGRP) was found to evoke the production of ARP1 in the presence or absence of insulin. In contrast, ARP2 was detected in both control and amylin-stimulated muscles. Amylin stimulation evoked incorporation of [35S]-methionine but not 32Pi into the protein and increased its concentration significantly. It is concluded that amylin elicits the production of ARP1 through phosphorylation and increases the protein biosynthesis of ARP2; the amylin-evoked production of ARP1 is insulin independent; amylin and CGRP share, at least in part, an intracellular signal transduction pathway; and ARP1 and 2 may be involved in the development of insulin resistance. It is suggested that ARP1 and 2 could potentially be used as molecular markers for the analysis of amylin action.

The work in this thesis concerns the analysis of essential replication genes of the F plasmid of Escherichia coli. The techniques principally employed were the cloning of restriction fragments, transposition mutagenesis, maxicell analysis of proteins, and DNA sequencing. This work led to the following findings. 1. A 1.15kb region of f5 was shown to contain all the functions required for initiation of F replication. Insertions of a transposon within this 1.15kb region abolished mini-F replication. 2. A mini-F plasmid was constructed which constitutes the smallest F derivative so far reported. This plasmid has an elevated copy number, as a result of deletion of the incC region. 3. At least two proteins specified by the essential F replication region were identified in maxicells. The promoter fragments of two mini-F genes were fused to a β-qalactosidase gene and demonstrated to have low but significant in vivo activities.

The Majority of Rotorua Lakes are rapidly becoming eutrophic and/or dominated by exotic hydrophytes. Lake Rotoma was studied to provide essential information on native aquatic hydrophytes within an oligotrophic lake, and to record the initial transition towards an exotic dominated community. The lake is of volcanic origin and has a markedly fluctuating water level because of the absence of a surface outlet. An underwater SCUBA investigation was made in 1972, followed by intensive sampling from 50 systematically located one transects in 1973. A suitable quadrat size and shape was selected and the sampling error was assessed. The species frequencies were calculated from presence or absence recordings. Quadrat cover, depth, gradient and substrate were also determined. Three characean species, Chara fibrosa f. acanthopitys. (A. Br.) R.D.W., Nitella leptostachys var. leonhardii (R.D.W.) R.D.W. and N.pseudoflabellata var. mucosa (Nordst.) Bailey, dominated the vegetation. The next most common pattern contained the same three species plus Lagarosiphon major (Ridley) Moss. The initial stages of L.major invasion in to a characean community was recorded from 1973 up to 1976. The charophytes ranged in depth from 0 - 17.5m and were located upon a wide variety of substrates and gradients. The native vascular plants were absent from many transects. The plants had a depth range from 0 - 4.5m with most occurring above 3.5m. The Low Mixed Community was primarily located in water less than 1.25ma at the N.E. end of the lake, which resulted in this area having a high species diversity. The exotic hydrophytes were more widespread and abundant than the native vascular phanerogams. The distribution of L.major and Elodea canadensis Michx. appeared to coincide with boating access, recreational usage and fallen submerged trees. The depth range was 0 - 6.0m, although the available habitats within this range have not yet been fully exploited. Emergent species were abundant within the S.W. inlet and also in the lagoons around the lake shoreline where sheltered conditions and shallow gradients prevail. An experimental study on the effect of erosion and sedimentation by removing the shallow water plant cover, showed that the presence of plants moderated the extent of erosion and also encouraged sedimentation. Recolonisation of cleared areas was discussed. Annual frequency changes in the vascular hydrophytes were largely influenced by water level fluctuations and by the extension of exotic species. Water-level fluctuations were observed to influence (a) the bottom limit of plant growth (probably indirectly through light penetration), (b) the upper limit of plant growth by determining available growing space, and (c) the rate of spread of exotic species by erosion and redistribution of existing stands. The 1973 raw data was processed and standardised in the form of species presence or absence, absolute frequency, and relative frequency within each transect. Distance matrices were constructed using non-Euclidean distance functions. A non-linear multidimensional scaling technique was used to construct transect and species ordinations. The advantages and problems associated with this technique are discussed and compared to alternative methods. That adopted was successfully tested for its ability to reconstruct a place map of the North Island of New Zealand. The configuration error was calculated for each ordination and was found to be less than for the alternative methods tested. From the first species ordination a concept of species prosperity was developed which involved the number of transects and quadrats within which each species was recorded, and the maximum depth recording for each species. Non-linear analysis indicated that average gradient might be a significant influence. From the first transect ordination, four vegetation groups were recognised and discussed. Depth, substrate, species dominance and gradient were significant within the ordinations. Non-linear analysis confirmed many of the previous interpretations of the ordinations but gave no further information.