Search

Climate change will drive substantial changes in water resources and uses across the world, while population growth, changing dietary preferences and biofuel use will require greater agricultural output. Yet farmland is not likely to increase, therefore yields must rise. Episodes of drought arising from an increasingly-unstable climate will also become more frequent, severe and persistent. As a result, water use in agricultural production must be managed both where water availability appears satisfactory and where water stress exists. Finding novel phenotypes which blend high yield and parsimonious water use could help water-efficient production, and two approaches were considered for the major cereal crop, barley. Circadian clock mutants introgressed into the elite variety, Bowman, were interrogated for their stomatal anatomical and physiological performance in relation to water use. In the short-term, Bowman A & gs peaked later, and WUEi earlier in the day than mutants under steady-state light, although dynamic responses to fluctuating light were not affected. Steady-state performance did not arise from anatomical causes, and were more likely to have a functional cause, where Bowman was sink-limited while clock mutants were source-limited. Drought of increasing severity was introduced to the clock mutants, and a mechanistic explanation of patterns of water use was sought in the production of the chloroplast-nuclear signalling molecule, phospho-adenosine phosphate (PAP). PAP concentration tends to be elevated in clock mutants compared to Bowman, and is elicited by drought stress. Clock mutants have better WUEi under drought than Bowman, and are more responsive to fluctuations in light, leading to improved relative yield. The water-use efficiency of a selection of wild barleys and landraces were also characterised relative to elite varieties. Stomatal anatomy was linked to gs, A and WUEi. Wild barleys and landraces sometimes produced better yield components (such as tillering) than elite varieties despite lacking targeted breeding. Meanwhile, phenotypic responses of the wild barleys were compared with the drought-tolerant variety, Bowman, under water stress conditions at germination and at tillering. Drought is an extreme form of WUEi risk, and water stress initiated a range behaviours in wild barleys that ranged between conservative and non-conservative but were seldom clear-cut in their overall phenotype. Drought affects stomatal density and size when initiated early in development. Physiological responses were more-important at tillering. Tiller number, leaf area and growth stage effectively differentiated between varieties, and all are valuable yield components that could be targeted by breeders.

In Botswana, the influx of male African elephants into the Makgadikgadi Pans National Park (MPNP) has resulted in the region becoming a hotspot for human-elephant interactions, with elephants leaving the MPNP to forage on crops. I sought to gain an understanding of human-elephant interactions in the region. None of the field characteristics studied indicated whether a field would be entered by elephants or the frequency of entry. Certain characteristics influenced the extent and value of damage. More isolated fields incurred larger areas of damage at the end of a field season and fields with a higher crop diversity resulted in larger areas of damage and a higher cost of damage. Crop-foraging events increased as the season progressed, with fewer events occurring during a full moon. Crop-foraging elephants did not adjust their group size outside the MPNP. However, the age of elephants predicted their probability of foraging on crops, with crop-foraging events predominantly involving older male elephants. Elephants showed directed movement towards, and foraged non-selectively but intensively within, fields. Negative experiences with elephants influenced farmers’ attitudes due to the occurrence of events, not the extent of damage. Farmers’ value for elephants was lower if elephants had entered their field that year and if they had encountered elephants that year, while tolerance was lower if elephants had entered their field that year and decreased with increasing numbers of crop-foraging events. Compensation estimates differed between stakeholder groups with farmers’ reporting the highest estimates of damage, followed by government estimates and then transect estimates, attributable to the structure of the compensation system and differing perceptions of damage. These results highlight the importance of understanding human-wildlife interactions at the individual field level, and the need to understand attitudes of farmers beyond the direct costs of crop damage if humans are to coexist with elephants.

Motor neuron disease is an incurable neurodegenerative condition. A proportion of motor neuron disease is inherited in autosomal dominant manner, of which approximately 10% is due to mutations in the enzyme superoxide dismutase 1 (SOD1). The mechanism by which mutant SOD1 selectively kills motor neurons is not understood. Motor neuron disease due to mutant SOD1 has been modelled by the SOD1 transgenic mouse which possesses a SOD^{G93A} transgene array. This mouse develops progressive motor neuron loss, resulting in premature death. The Legs at odd angles (Loa) mouse, which has a missense mutation in the gene encoding cytoplasmic dunein, develops progressive – wild-type, Loa/+, SOD1 ^{G93A} and Loa/SOD1^{G93A}. Surprisingly, Loa/SOD1^{G93A} mice survive significantly longer than their SOD1^{G93A} littermates, thus the presence of a mutation in dynein reduces the toxic effect of mutant SOD1. To gain further understanding of how mutant SOD1 kills motor neurons, the behaviour of mutant SOD1 between neurons of these genotypes was compared. Differences in behaviour of mutant SOD1 between these genotypes may correlate with the observed differences in survival and thereby suggest toxic mechanisms. Methods of tagging SOD1 for live cell imaging were assessed and strategies of transfecting cultured motor neurons with tagged SOD1 evaluated. Lentiviral vectors were used to transduce cultured motor neurons with tagged SOD1. Methods of assessing anterograde axonal transport were evaluated. The localisation and movement characteristics of wild-type and mutant SOD1 were evaluated in wild-type motor neurons. No statistically significant differences were found. The localisation and movement characteristics of mutant SOD1 was evaluated in motor neurons of genotypes wild-type, Loa/+, SOD1^{G93A} and Loa/ SOD1^{G93A}. No statistically significant differences were found. Thus the behaviour of the toxic moiety, mutant SOD1, was not demonstrated to be different between neurons with different survival characteristics.

This thesis contains results from a long-term campaign to monitor the behaviour of Be/X-ray binary systems at multiple wavelengths. The work presented in this thesis concentrates on the application of Be/X-ray binary observations to contribute to the understanding of the Be star phenomenon itself. I present studies of the X-ray behaviours of these systems - effectively using the Neutron Star as a probe of the Be star's envelope - and seek correlation between this and the behaviour of the Be star inferred from optical and infrared observations. I attempt to model the X-ray light-curves as accretion from a circumstellar disc, and investigate the validity of these techniques by considering recent claims that the orbiting neutron stars in Be/X-ray binaries perturb the Be stars' circumstellar discs. Finally, I present a study of Be/X-ray binaries in the Magellanic Clouds, identifying optical counterparts to a number of recently discovered systems.