Sixteen native shrub species with various breeding systems and pollination syndromes were investigated in geographically separated populations to determine breeding systems, reproductive success, population structure, and habitat characteristics. Of the sixteen species, seven are hermaphroditic, seven dioecious, and two gynodioecious. Two of the dioecious species are cryptically dioecious, producing what appear to be perfect, hermaphroditic flowers, but that function as either male or female. One of the study species, Raukaua anomalus, was thought to be dioecious, but proved to be hermaphroditic. Teucridium parvifolium, was thought to be hermaphroditic, but some populations are gynodioecious. There was variation in self-compatibility among the four Alseuosmia species; two are self-compatible and two are self-incompatible. Self-incompatibility was consistent amongst individuals only in A. quercifolia at both study sites, whereas individuals in A. macrophylla ranged from highly self-incompatible to self-compatible amongst four study sites. The remainder of the hermaphroditic study species are self-compatible. Five of the species appear to have dual pollination syndromes, e.g., bird-moth, wind-insect, wind-animal. High levels of pollen limitation were identified in three species at four of the 34 study sites. Moderate to high levels of pollen limitation were evident in 50% of the gender dimorphic populations compared with 31% of hermaphroditic populations. Melicope simplex populations were female-biased in 14 study plots and successful fruit set was influenced by combinations of male plant density, male flower numbers and distance to the nearest pollen. Natural fruit set in wind-pollinated species was generally higher than in animal-pollinated taxa. Populations of Coprosma spathulata were mostly malebiased, and male plant density and proximity influenced natural fruit set, with a correlation between low fruit set and low male density. Population recruitment was evident in 32 of the 34 sites. In Pimelea arenaria, recruitment failure was widespread in populations throughout the North Island despite high seed set. Gender ratios showed variation from north to south, with a lower proportion of females with increasing latitude. Many of the sixteen shrubs in this study show remarkable resilience to the effects of fragmentation of natural ecosystems, and many of the species have benefited from the creation of new edge habitat after perturbation - it may be that edges are important refugia for some native taxa, especially shrubs.

To produce plant material for this thesis, Trifolium repens (white clover) (genotype 10F) and Trifolium occidentale (genotype 18Z) were propagated to produce individual stolons trained over a plastic matrix to inhibit nodal root formation. These stolons comprised leaf tissue representative of all developmental stages, from leaf initiation, maturation through to senescence. The developmental pattern for both species in terms of leaf ontogeny was generally reproducible between vegetatively propagated clones. Three distinct 1-aminocyclopropane-1-carboxylatc (ACC) oxidase genes expressed during leaf ontogeny in white clover (Trifolium repens L.) have been identified (Hunter et al., 1999). Of the three ACC oxidase genes identified, one designated TR-ACO2 is expressed in newly initiated and mature green leaves while TR-ACO3 is expressed predominantly in the senescent leaf tissue. In order to further characterize the protein products of these genes, a series of FPLC columns was used to partially purify isoforms of ACC oxidase from leaf tissue of white clover at different developmental stages, followed by 2D gel electrophoresis to obtain further purification. Two distinct isoforms of ACC oxidase were identified and partially purified from newly initiated green (designated the NIGI isoform) and senescent (designated the SEI isoform) leaf tissue. Both purified NIGI and SEI proteins were recognized by western analysis using an anti-(Trifolium repens) TR-ACO2 antibody after SDS-PAGE or 2D gel electrophoresis. To determine whether NIGI is coded for by TR-ACO2 and SEI is coded for by gene TR-ACO3, protein spots (after 2D gel electrophoresis) were digested with trypsin and the masses of the peptide determined by matrix-assisted laser desorption ionization-timc of flight (MALDI-TOF) mass spectrometric analysis. For NIGI, the coverage of the putative protein sequence (TR-ACO2) by tryptic digestion ranged from 24.5% to 37.6%, while the observed pI (5.1) and molecular mass (37 kDa) were close to the theoretical pI (5.3) and computed mass (35.7 kDa). For SEI, the percentage coverage of the putative protein sequence (TR-ACO3) from the peptides identified ranged from 13.4% to 18.0%, while the observed pI (5.2) and molecular masses (35.0-35.5 kDa) were also close to the theoretical pI (5.5) and computed mass (35.2 kDa). These data suggest that the NIGI isoform is encoded by TR-ACO2, while the SEI isoform is encoded by TR-ACO3. ACC oxidase activity in vitro and ACC oxidase protein accumulation over 24 h in mature green leaf tissue extracts during both short and long days has been shown to be under circadian control. There are two ACC oxidase activity peaks observed, in which the pattern of fluctuation in ACC oxidase activity resulted in a high level of enzyme activity at 12:00 am (0.18-0.27 nmol ethylene/h/mg), and maximum activity at 12:00 pm (0.24-031 nmol ethylene/h/mg). Lowest activity was observed in both long and short days at 9:00 pm (0.09-0.10 nmol ethylene/h/mg). In addition, northern analysis indicated that the TR-ACO2 mRNA level also displayed a circadian pattern of expression. Investigation of the effect of protein phosphorylation and dephosphorylation on ACC oxidase activity indicated that ACC oxidase activity in vitro during the periods of maximum activity increased 36% (at 12:00 am) and 56% (at 12:00 pm) after dephosphorylation, respectively. However, there was only 21% increase in enzyme activity at the time point with lowest activity (9:00 pm) in the dephosphorylated extracts. SDS-PAGE using a mini-protein gel system or a gradient gel system showed that the molecular mass of ACC oxidase decreased after dephosphorylation when compared with phosphorylation of the enzyme. These results suggest that the phosphorylation and dephosphorylation of the ACC oxidase proteins occurs in vitro and the state does affect enzyme activity. In the second part of this thesis, the coding regions of putative ACC oxidase gene transcripts were generated from leaf tissue of genotype 18Z of T. occidentale using RT-PCR. Sequence alignments indicated that the sequences could be grouped into two distinct classes, and these coding regions were designated TO-ACO2 (Trifolium occidentale ACC oxidase 2) and TO-ACO3 (Trifolium occidentale ACC oxidase 3). TO-ACO2 and TO-ACO3 share 82% similarity in nucleotide sequence and 84% similarity in amino acid sequence. The TO-ACO2 and TO-ACO3 sequences were validated as encoding ACC oxidases by comparison with other ACC oxidases in the GenBank database and both TO-ACO2 and TO-ACO3 deduced amino acid sequences contain all the residues hitherto shown to be important for maximal activity of the enzyme. Further, TO-ACO2 had 97% identity with TR-ACO2 at the nucleotide level, and 98% identity at the amino acid level. TO-ACO3 had 97% identity with TR-ACO3 at the nucleotide level, and 96% identity at the amino acid level. Genomic Southern analysis, using 3'-UTRs of TR-ACO2 and TR-ACO3 as probes, could not confirm that TO-ACO2 and TO-ACO3 arc encoded for by distinct genes. Expression studies of TO-ACO2 and TO-ACO3 genes during leaf maturation and senescence of T. occidentale were examined using northern analysis. TO-ACO2 is expressed predominantly in newly initiated and at the onset of the mature-green leaf stage, while TO-ACO3 shows maximal expression in senescent leaf tissue. The changes of ACC oxidase activity during leaf ontogeny of T. occidentale coincided with the pattern observed for ACC oxidase protein accumulation using western analysis and image analysis.

A field experiment was conducted to examine the responses in growth, total dry matter (TDM), seed yield and nitrogen (N) accumulation of Kabuli chickpea cv. Principe and narrow-leafed lupin cv. Fest to different irrigation levels and N fertilizer on a Templeton silt loam soil at Lincoln University, Canterbury, New Zealand in 2007/08. The irrigation and fertilizer treatments were double full irrigation, full irrigation, half irrigation and nil irrigation and a control, full irrigation plus 150 kg N ha⁻¹. There was a 51 % increase in the weighed mean absolute growth rate (WMAGR) by full irrigation over no irrigation. The maximum growth rates (MGR) followed a similar response. The growth rates were not significantly decreased by double irrigation. Further, N fertilizer did not significantly improve crop growth rates. With full irrigation MGRs were 27.6 and 34.1 g m⁻² day⁻¹ for Kabuli chickpea and narrow-leafed lupin, respectively. Seed yields of fully-irrigated crops were trebled over the nil irrigation treatment. With full irrigation, seed yield of chickpea was 326 and that of lupin was 581 g m⁻². Seed yield of the two legumes was reduced by 45 % with double irrigation compared with full irrigation. Nitrogen fertilizer did not increase seed yields in either legume. Increased seed yield with full irrigation was related to increased DM, and crop growth rates, seeds pod⁻¹ and seeds m⁻². Crop harvest index (CHI) was significantly (P < 0.05) increased by irrigation and was related to seed yield only in narrow-leafed lupin. With full irrigation, the crops intercepted more than 95 % of incoming incident radiation at leaf area indices (LAIs), 2.9 and 3 or greater in Kabuli chickpea and narrow-leafed lupin, respectively. In contrast, without irrigation the two legumes achieved a maximum fraction of radiation intercepted of less than 90 %. With full irrigation, total intercepted photosynthetically active radiation (PAR) was increased by 28 % and 33 % over no irrigation for Kabuli chickpea and narrow-leafed lupin, respectively. Fully-irrigated Kabuli chickpea intercepted a total amount of PAR of 807 MJ m⁻² and fully-irrigated narrow-leafed lupin intercepted 1,042 MJ m⁻². Accumulated DM was strongly related to accumulated intercepted PAR (R² ≥ 0.96**). The final RUE was significantly (P < 0.001) increased by irrigation. With full irrigation the final RUE of Kabuli chickpea was 1.49 g DM MJ⁻¹ PAR and that of narrow-leafed lupin was 2.17 g DM MJ⁻¹ PAR. Total N accumulation of Kabuli chickpea was not significantly affected by irrigation level. Kabuli chickpea total N was increased by 90 % by N fertilizer compared to fully-irrigated Kabuli chickpea which produced 17.7 g N m⁻². In contrast, total N accumulated in narrow-leafed lupin was not increased by N fertilizer but was decreased by 75 % with no irrigation and by 25 % with double irrigation (water logging) compared to full irrigation with a total N of 45.9 g m⁻². Total N was highly significantly related to TDM (R² = 0.78** for Kabuli chickpea and R² = 0.99** for narrow-leafed lupin). Nitrogen accumulation efficiency (NAE) of narrow-leafed lupin was not affected by irrigation or by N fertilizer. However, the NAE of Kabuli chickpea ranged from 0.013 (full irrigation) to 0.020 (no irrigation) and 0.017 g N g⁻¹ DM (full irrigation with N fertilizer). The N harvest index (NHI) was not affected by irrigation, N fertilizer or legume species. The NHI of Kabuli chickpea was 0.50 and that of narrow-leafed lupin was 0.51. The NHI was significantly (r ≥ 0.95 **) related to CHI.

The response of cucumber (Cucumis sativus L.) plants to various root and shoot environments (solution depth, temperature, ionic strength, nitrogen and calcium level and light intensity) were studied. Cucumber plants were grown in continuously circulating-solution in a heated-glasshouse. Dry weights of leaves, stems and roots, leaf area, leaf number, root length and root number were measured as well as uptake of potassium, calcium and nitrogen. The relationship between shoot and root in terms of functional equilibrium equations was also examined. The results presented show that: 1. Shoot growth of cucumber plants was reduced if grown in solutions of less than 50mm in depth; 2. When roots were grown in shallow solution depths at 1 or 5mm the dry weight allocated to the root increased. The ratio of root number/root length(no./cm) also increased. Lowering solution temperature to 12.5±2.5°C enhanced the production of root number relctive to root length, and 5 and 2% of full strength and 5% of full strength nitrogen level solution stimulated the growth of root length relative to root number; 3. Under low solution temperature treatment leaf number was maintained at the expense of leaf area. Under low total ionic strength and low nitrogen solution, enhanced root length growth was at the expense of leaf area growth; 4. Low solution temperature enhanced the dry weight allocated to the stem relative to the leaf. Low total ionic strength and low nitrogen solution increased the dry weight allocated to the leaf relative to the stem; 5. The specific activity of root, represented by specific absorption rate, increased when the shoot was under light stress and, the specific activity of shoot, represented by unit shoot rate, increased when the root was under nitrogen-stress; 6. The form of equation developed by Thornley (∆M = fm∆W, where ∆M is the increment in weight of element M and ∆W the increment in total plant dry weight during a time period ∆t with fm a constant) showed a better relationship than the equation developed by Davidson [root mass x rate(absorption) ∝ leaf mass x rate(photosynthesis)] and subsequently used by Hunt in the form of mass ratio(root/shoot) ∝ l/activity ratio; 7. The equation developed by Chung et al, total plant weight/(leaf number/leaf area) ∝ total "k"/(root number/root length), where k represents the total contents of elements or compounds, showed a good approximation of the relationship between shoot and root under all the environmental stresses imposed with the exception of calcium uptake. The results support the concept that the activity of the root or shoot in carrying out its function is influenced by the demand created by the opposite organ and appears to be a better assumption than that which proposes that the activity of an organ is solely dependent on its own size.