A controlled experiment was established in primary upper montane rain forest (UMRF) in the Blue Mountains of Jamaica to investigate the ecophysiological response of seedlings of seven tree species to canopy gap formation. The group included one invasive species, the Australian tree <I>Pittosporum undulatum,</I> with a mixture of native and endemic shade tolerant and intolerant species. The main aim was to assess whether the known approximate light demand of seedlings of the species for regeneration correlated to the relative ecophysiological performance of the seedlings across a light gradient. Gaps were created artificially by de-branching of canopy trees, ensuring minimal possible damage to the seedling bank. Resulting light levels were equivalent to those associated with damage from a major hurricane. The photosynthetic characteristics, morphology and longevity of shade leaves and seedlings were investigated in naturally growing seedlings on four gap and four understorey control plots. Above- and below-ground competition and the effect of leaf masking and removal were investigated in separate sub-experiments using potted seedlings. In the understorey, mean levels of total diurnal photosynthetic photon flux density (PPFD) on the uppermost leaf were 2.14% of diurnal integrated above canopy levels. No significant variation occurred among the sevens species, whether defined by instantaneous diffuse light (under cloud), or the proportion of cloud-free time for which sunflecks were experienced. The three indices of light level showed strong, though not always 1:1, interrelationships. The findings give strong ecophysiological backing to existing knowledge of the species' regenerational ecology across a range of disturbed and steady state environments in UMRF. They also give information as to relative physiological performance of native species and the problematic invasive tree <I>Pittosporum undulatum. </I>

Coniferous tree species growing on sites exposed to aerial pollution exhibit an altered pattern of vegetative development compared to those on unpolluted sites and contain higher needle cytokinin contents. This study aimed to determine which pollutants promoted the enhanced cytokinin status by analysing the needle cytokinin content of Sitka spruce (<I>Picea sitchensis</I> (Bong) Carr) trees sprayed with individual components of aerial pollution. Only needles of neutral N treated trees contained significantly higher isoprenoid cytokinin contents than that of control trees sprayed with water. In contrast, needles of trees treated with acid-mists containing S and N contained significantly lower levels of cytokinin than controls, indicating that S pollutants, or low pH, or both, significantly suppress the nitrogen-mediated stimulation of cytokinin levels. Changes in cytokinin metabolism were also recorded. Several aromatic cytokinins were detected in the needles of Sitka spruce for the first time. In contrast to the isoprenoid cytokinins, the aromatic cytokinin content was significantly reduced in needles from N treated trees. The function of this enigmatic group is unknown, although possible roles for these cytokinins in Sitka spruce are discussed. Beech seedlings (<I>Fagus sylvatica</I> L.) fumigated with NH<SUB>3</SUB> also contained significantly higher cytokinin contents in the roots compared to controls fumigated with charcoal-filtered air. Changes in cytokinin metabolism were also detected. Defoliation and needle yellowing are common symptoms of trees growing on polluted sites. Sitka spruce seedlings were partially defoliated with the aim of characterising the changes in cytokinins resulting specifically from needle loss. Needles of defoliated plants contained significantly higher levels of O-glucosyl isoprenoid cytokinins, a 'diagnostic' by which the influence of defoliation might be recognised in the field.

I studied the Blue &amp; John Crow Mountain National Park in Jamaica to evaluate protected area effectiveness in reducing forest clearance in this country, which is a significant contributor to the high biodiversity and high endemism of the Caribbean region. Using time-series Landsat imagery, I found that forest clearance occurred at a similar high rate before and after the declaration of the area as a national park (1.50 % yr^-1 and 1.42% yr^-1 respectively), and that forest clearance and fragmentation were mostly confined to the more accessible forest at lower elevation (&lt; 1000 m). In these lower elevation forests, I documented high plant endemism (30 %) and beta diversity using data from 25 x 0.04 ha plots, which I established and Adams (1972) flora of Jamaica. To investigate the conservation value of secondary forest, I used the oldest secondary forest ever compared with adjacent old growth forest to assess tree species composition and structure in the two forest types. I showed that although these two forest types were similar in many aspects, they were still different in the percentage of tree individuals that are endemic, and in the species composition of the larger stems. In a more controlled experiment, using one of the longest monitored sets of permanent plots in the tropics, I showed that 24-35 years after intensive disturbance in experimental gaps, the species compositional and basal area of trees in gap plots had not recovered to their pre-cut condition, and gaps were heavily invaded by a non-native species. My results indicate the slow recovery from disturbance of tropical montane forests, and the persistent effect of historical disturbance on secondary forests.

This thesis suggests that agriculture has become locked-in to a particular way of growing the food we eat. It is argued that alternative technologies and techniques exist which are, arguably, less damaging in environmental terms, and which need not incur unacceptable yield penalties. It is suggested that the increasingly important linchpin of this way of doing agriculture is the seed, in particular, genetically uniform elite varieties bred for uniformity. A critique of orthodox theories of technological change in agriculture, including those of Boserup and Hayami and Ruttan, is offered before undertaking an exploration of other concepts which have emerged from recent work in the study of science and technology. An attempt is made to develop an unorthodox framework for understanding the development of technical change in agriculture by drawing on these concepts. Three case studies are offered, those of hybrid maize in the United States, the early days of the Green Revolution in Mexico, and the changes, especially in the realm of intellectual property rights legislation, brought forward to foster the development of new biotechniques. In each case, it is shown that alternatives ways of doing agriculture could have been developed, but for one or other reason, the development of research was such that one path, the one dependent on genetic uniformity, was chosen. In concluding, the thesis notes the problems associated with trying to change the prevailing mode of practising agriculture which appears to be locked-in to genetic uniformity.

This thesis set out to investigate the light responses of shade-tolerant tropical tree species, and to assess whether or not there was a trade-off between shade tolerance and responsiveness to high light. Shadehouse experiments indicated that low-light relative growth rate was positively correlated with that of high light across species, and all seedlings showed typical 'sun' and 'shade' characteristics when grown in high- and low-light survival amongst the species studied. These trends were mirrored in forest-growth plants. Seedling mortality rate in the forest varied across species and was driven by vertebrate herbivory rather than light availability. In the shadehouse studies, seedling light responses did not appear to be confounded by nutrient limitation. Some evidence was found for phylogenetic effects across species, particularly in root morphology. Ontogenetic effects were also found for a number of plant traits, although the significance of these effects was not consistent across species. Species showed changes in growth ranking both across light levels and over time, suggesting a role for shifting competitive hierarchy in promoting species coexistence. The shade-tolerant species studied also varied significantly in the flexibility of their plant traits in response to changes in light conditions. Nonetheless, all species showed significant acclimation to an increase in light in terms of growth, photosynthesis, morphology and architecture. Acclimation to a decrease in light involved a significant decrease in photosynthetic rate and growth across species, but changes in morphology were limited.

New Zealand has no native mammalian herbivores, and the recent introduction of deer to the country is having a major impact on native vegetation. This thesis explores aspects of the foraging ecology of introduced deer and the responses of native plant species to browsing. In Chapter 1, I introduce the study system and the conceptual themes of the thesis, which run along two main axes. The first theme relates to the importance of traits of individuals (both plants and deer) versus landscape or community-level characteristics in determining diet selection. The second theme is concerned with two mechanisms by which native plant species can cope with ungulate herbivores: avoidance versus tolerance of damage. In Chapter 2, I use a pre-existing vegetation and browsing survey dataset to explore the capacity of 53 forest plant species to avoid damage. At a community level, plants were more heavily browsed if their neighbours were of high palatability. At the level of the individual, the strongest predictor of the palatability of species was the dichotomy between woody and non-woody plants (mostly ferns). Among woody plant species, the primary predictor of palatability was low content of phenolics. There were weaker associations between palatability and high SLA, large leaf size, and low lignin, but the significance of these weaker traits depended on the index by which palatability was defined. The considerably lower palatability of ferns compared with woody plants could not be explained by foliar traits measured in this study. Differences in palatability between fern species were small, and not related to any foliar trait. In Chapter 3, I use a pre-existing rumen contents dataset to explore major sources of variation in diet.

No description available.

No description available.

No description available.

No description available.