Two transects across the hybrid zone between the fire bellied toads, <i>Bombina bombina </i> and <i>Bombina variegata</i> were studied. The first, near Pescenica in Croatia, had been the subject of a number of previous studies. The second, at Crna Mlaka, covered a wider region and its geography was more diverse. This thesis presents results of the first study to be undertaken in this part of the <i>Bombina</i> hybrid zone. Using data from the Pescenica transect, the cline for a single locus allomorphic enzyme trait was compared to that of the multi-locus trait for ventral colouration. A selection of ecological variables, extracted from a satellite image and a set of topographical maps, were examined for their effect on the position and width of the clines. The results were similar for both clines: the measure of forest cover extracted from the satellite image was found to give a significant improvement to the likelihood model of clines for both traits, with the multi-locus trait appearing to be more sensitive to ecological variation. The hybrid zone in the Crna Mlaka region was described and the position of a one dimensional cline for the ventral colouration trait estimated. The effect of ecological parameters altitude, contour density and forest cover were assessed, and it was established that the measure of contour density significantly improved the fitted model. This finding is in accordance with the hypothesis that movements of <i>Bombina bombina</i> are limited by the relief of the landscape rather than the absolute altitude.

Oak cynipids initiate the growth of highly complex galls on plants in the genus <i>Quercus </i>and related genera in the family Fagaceae. These galls support closed communities consisting of high diversities of gall inducers, inquilines and natural enemies. Key questions regarding the evolution and ecology of these communities remain unanswered. In particular, the role that phylogeography of component species plays remains unclear. Communities associated with different gall wasps vary in composition and structure, but little is known about the processes generating such variation. This thesis addresses the question of the ‘community phylogeography’ of the oak wasp system, using tree-based Bayesian analyses of mitochondrial sequence data. Work already done on the gall wasp hosts is extended to the next trophic level through examination of the European phylogeography of several chalcid parasitoids. The resulting patterns are examined for concordance between parasitoids and the gall wasps, and hypotheses regarding ancient and more recent range expansion are evaluated. The phylogeny of the oak gall wasp attacking Torymidae one of the major families of parasidoid attach oak cynipids, is estimated and hypotheses regarding the evolution of the group are considered. In addition, the use of cytochrome oxidase barcodes as a means of data collection from oak gall wasp communities is evaluated. Current methods are time consuming, labour intensive, and often error prone and an efficient standardised data collection procedure would be beneficial to further analyses.

This problem of fish-eating birds was widespread at Scottish fish farms. Grey heron (<i>Ardea cinerea</i>), cormorant (<i>Phalacrocorax carbo</i>), shag (<i>Phalacrocorax aristotelis</i>), goosander (<i>Mergus merganser</i>) and red-breasted merganser (<i>Mergus serrator</i>), the principal species implicated, were studied in an area on the west coast of Scotland from September 1985 to August 1987. Data were also collected from farms in south Argyll, Highland and Tayside Regions. Herons visited both pond farms, taking fish in the shallows, and cages, taking fish through top nets. Fish were either eaten or dropped, the resulting wounds often making them unmarketable. Birds tended to arrive at cages during the hours of darkness, selecting those cages containing fish weighing less than 300g and taking fish swimming close to the surface. These fish were generally smaller than the majority held in the cage. Adults were aggressive towards first-years who had a lower feeding success. However, both age classes increased their food intake rate by feeding at fish farms. In some cases this increased food supply appeared to influence breeding, allowing birds to lay earlier, produce larger clutches and larger broods surviving to fledge. Such increased production was offset by widespread persecution. Between 1069 and 1936 birds were estimated to be killed annually, of which c. 14% were adult, but it was considered unlikely that this reduced the overall Scottish population. Cormorants visited marine and freshwater cage farms, particularly during the winter. In freshwaters they may have been attracted by large numbers of escaped rainbow trout living in the waters adjacent to cages. Underwater attacks could cause considerable damage to fish in cages, but birds did not remove fish from them. Anti-predator nets reduced this damage, but were not toally effective. Between 980 and 3047 birds were estimated to be killed annually, and as 25% were adult, the Scottish population could have been affected. Shags may have been attracted to marine cages by the large concentrations of wild fish living in adjacent waters. Juveniles, but not adults, had an increased food intake rate fishing close to farm cages when compared with elsewhere. There was almost no evidence of damage to stock and persecution was therefore unwarranted. Between 1930 and 4255 birds were estimated to be killed annually, but almost all were juveniles and it was unlikely that such pesecution reduced the Scottish population. Goosanders were a problem at a few sites during the spring. Gulls and crows scavenged dead fish and fish food opportunistically, while other species which visited farms caused no problems. Losses to fish-eating birds were small in comparison to other forms of fish mortality and represented only a small fraction of a farm's running costs.

No description available.

The distribution of species of caddisfly larvae within the study stretch of the Whiteadder Water was attributed to three main environmental factors, near bed velocity, depth and amount of Coarse Particulate Organic Matter and was consistent with previous research. Species of caddisfly that occurred within riffles generally had unsynchronised lifecycles, compared to a more synchronised lifecycle to species that occurred in pools, and this was the most pronounced in the case-less caddisfly larvae. The incidence of stranding in cased caddisfly larvae n bank side quadrats increased with discharge once a threshold had been exceeded. The incidence of stranding was dependent on the adjacent bank side microhabitat, topography and larval case type. Stranding was not homogeneous, with more individuals stranded adjacent to riffles than pools on gravel substrates than turf substrates. Cased caddisfly larvae with fine mineral cases mineral cases were the most abundant type found stranded, while species with vegetative cases were rarely found, though present in the benthos. Flume experiments indicated that cased caddisfly larvae were passively deposited on the inundated bank sides and they did not actively seek this refuge. The ability of some cased caddisfly larvae to regain the water once stranded varied with case type, and the topography (substrate and slope) of the bank in experimental arenas. Time taken to regain the water was species/case specific with vegetative cased larvae regaining the water fastest and fine mineral cased larvae the slowest. Time taken to regain the water increased with increasing substrate complexity and distance from the water, but decreased with increasing slope. The incidence of stranding of cased caddisfly larvae by hydraulic disturbance within the Whiteadder Water, while it may be important did not have a significant negative impact on the assemblage of cased caddisfly larvae. Only an estimated 12% of the vulnerable population of cased caddisfly larvae are lost from the benthic population by stranding over a year with a typical discharge regime.

No description available.

Genetic polymorphism of <I>Plasmodium falciparum</I> populations has been studied in two localities in Sudan, Khartoum the capital and Asar village in the eastern region. The frequencies of alleles of three antigens denoted MSP-1, and MSP-2 and Exp-1, three enzymes and nine polymorphic proteins detected by 2D-PAGE were investigated. Parasites from both areas were found to be very diverse, and frequently patients were infected with more than one genotype. Allelic polymorphism was slightly lower in Asar than in Khartoum. <I>In vitro</I> responses to chloroquine, pyrimethamine and mefloquine were assessed among the parasites of Asar in 1989 and 1990. The parasites were found to have diverse responses to chloroquine and pyrimethamine in both years. The prevalence of parasites sensitive and resistant to pyrimethamine varied considerably between 1989 and 1990, but the responses to chloroquine remained more or less stable. Characterisation of R1-type chloroquine resistant parasites for antigen genes proved that they were genuine recrudescent resistant parasites, and not derived from new infections. The antigens, 2D-PAGE proteins and enzymes, were again monitored in 1990. In addition, alleles of MSP-1 and MSP-2 were characterised by PCR-amplification and allele-specific DNA probes in 1990 and 1991. Some changes in the frequencies of certain alleles were noted from one year to another. The MSP-1 & MSP-2 alleles identified by the PCR/DNA probes and the frequency of multiclonal infection in Asar in 1991 were compared to those of parasites collected in the same year in a village of similar size, Michenga, in Tanzania.

The major aim of this thesis is to develop, apply, and test resource competition theory in the ungulate communities of Africa. Ungulates were chosen because RCT requires a great deal of information about the foraging ecology of the species to be modelled, and foraging ecology in ungulates has probably been subjected to more detailed research than any other animal group. The first chapter reviews the literature on competition in ungulates. Chapter 2 then develops a description of ungulate resources within the RCT framework. This new description efficiently encapsulates many well known features of ungulate foraging ecology. Several ungulate foraging scenarios are then described within the RCT framework in a series of conceptual models. A strong prediction that stable co-existence is possible between species that are sufficiently, but not too much, different in body size, emerges. One of these models, describing competition between topi and wildebeest, is developed more fully and tested using data from the literature. The model predicts the resource supply conditions under which each species can be expected to exclude the other. Co-existence is shown to be promoted by environmental heterogeneity that is generated by the foraging activity of the animals themselves. The modelling approach is developed in chapter 3, with a generic model of resource competition between two species differing in body size. The effect of body size on energetic requirements, intake rate, digestive capacity, and selective ability are integrated within the model. Despite the large literature on ungulate foraging ecology, there remain areas of insufficient knowledge, in particular a quantitative description of the relationship between body size and selective ability. The model predicts that stable co-existence is possible between two species competing for resources.

No description available.

This thesis examines the genetic and ecological consequences of the introduction of exotic Japanese sika deer (<I>Cervus nippon nippon</I> Temminck. 1838) to the range of congeneric red deer (<I>Cervus elaphus scoticus</I> L. 1758) in Scotland. Sika were introduced as result of escapes and deliberate releases of park deer around the turn of the century and have since established feral populations in several locations. Most of these populations are achieving high reproductive rates, are currently expanding their range and are hybridising with red deer. These circumstances are of practical concern as red deer have considerable economic importance in Scotland, both for the positive values of stag stalking and leisure use of the hills and through the negative effect of damage to crops and forestry. The situation is also of biological importance as it provides a rare opportunity to assess the genetic structure of a hybrid population after an introduction and to examine the functional significance of hybridisation affecting quantitative traits. Sampling transects were established across two sympatric sika-red population ranges, in Argyll and Invernesshire. Genotypes of 235 deer from nine forests in Argyll and 136 deer from 7 forests in the Great Glen, Invernesshire, were scored at 2 isozyme, 2 microsatellite and 3 mitochondrial DNA (mtDNA) marker loci. The species-specific mtDNA markers were generated within this study. Clines in sika allele frequency were found across both transects, with sika frequency reaching 0.8 in Argyll and fixation (1.0) in Invernesshire. Sika are assumed to have at least one and probably several selectively advantaged genes to account for this consistent and rapid increase in frequency. There were significant heterozygote deficits and linkage disequilibria in the centre of the clines, even though samples in the Great Glen were small. MtDNA clines were steeper than nuclear clines, indicating the role of sika or hybrid stags in propagating the advance, but patterns of cytonuclear disequilibrium were not informative about mating pattern, due to the possibility of hybrid immigration and the influence of heterozygote deficit reducing sample sizes. There was no evidence of directionality in the F1 cross, though assortative mating in the backcross could contribute to the linkage disequilibrium observed, as could selection against hybrids.