Using SCUBA and an in situ method of quantifying substrate characteristics, I describe patterns of zebra mussel (Dreissena polymorpha) distribution along the St. Lawrence and Hudson Rivers and in Oneida Lake, New York, and develop empirical models for their abundance. Calcium-poor waters originating from rivers draining the Canadian Shield resulted in a lack of zebra mussel along the north shore of the St. Lawrence River east of Montreal until Portneuf despite an abundance of suitable substrate. Calcium concentrations of 15 mg/L or less were found to limit the distribution of zebra mussel. The entire south shore from Cornwall, Ontario to Ile d'Orleans, Quebec was colonized by zebra mussel wherever suitable substrate was found. In the Hudson River, along the south shore of the St. Lawrence River and in Oneida Lake variability in density was primarily related to substrate type which explained between 38% and 91% of the variance. Other factors such as Secchi depth, calcium concentration of the water, the presence of crayfish, native unionid abundance and the maximum width of the river at the site increased the amount of explained variance across the different systems. The influence of substrate type on zebra mussel density and the predictions of the model were also tested using data from the literature, where substrate type explained 75% of the variability in density. The scatter of the literature data above the predictions of the empirical model suggests that North American zebra mussel populations may continue to grow before reaching equilibrium levels. While water chemistry parameters may be useful predictors of the presence or absence of zebra mussel in a given water body, physical factors play a far greater role in determining local abundance.

Individuals within a species often compete for resources in both space and time. In dominance hierarchies individuals with the greatest competitive ability will occupy prime locations during optimal periods to increase efficiency in gathering a resource. Subdominant individuals with low competitive abilities may be forced to reside in habitats of poor quality relative to dominant individuals. In this study I examined the long term patterns of giant kokopu distribution, movement, growth, habitat use, and social interaction between fish in relation to invertebrate drift (food supply). The habitat quality and abundance of food along a one km section of Alex�s Creek was monitored for a two year period. Though the physical structure of Alex�s Creek was relatively homogenous there were significant spatial differences in the density of drifting invertebrates sampled between riffles and pools over this long temporal period. In general, more drifting invertebrates were sampled in riffles of relatively long length and area. Within Alex�s Creek the distribution of giant kokopu, Galaxias argenteus, was determined by patchy distributions of food supply, specific physical factors of pools, and interactions between fish in dominance hierarchies. Over the 20 month study period, the most important biotic factor determining fish biomass was the total number of drifting invertebrates within pools while the most important abiotic factor determining fish abundances within pools was the pool size (surface area). The growth rate of individual fish correspondingly varied between fish residing within pools of different quality. Growth rates were higher for individual fish residing in pools with a relatively high density of drifting invertebrates. The differences in individual growth rates of giant kokopu may determine when fish leave a particular habitat patch (pool) and move to a new one. Fish that moved had lower growth rates (before moving) than fish that remained resident within home pools. Overall the movement of giant kokopu within Alex�s Creek was very restricted due to a relatively consistent distribution of food, however there were differences in the mobility of giant kokopu among different social ranks. The most dominant fish in pools were largely sedentary while fish ranked directly below fish 1 (i.e. fish 2 and 3 in social hierarchies) were relatively mobile. The ability of dominant fish to exclude subdominant fish from the most preferred feeding positions during optimal feeding times had consequences for overall giant kokopu activity. Under normal food supply conditions dominant fish were predominantly nocturnal and maintained large home ranges at night. Conversely subdominant fish occupied large home ranges by day, but were generally not observed at night. When the food supply was limited the utilization of pools was determined by social rank. Dominant fish from each pool increased daytime activity, home range size, aggressiveness, and the capture of food items offered by day while simultaneously reducing the spatial and temporal activity and habitat use of subdominant fish. These results indicate that behavioral changes in large dominant fish influence and reduce the amount of resources available to subdominant fish.

Stream ecology is dominated by studies that describe the distribution of invertebrate populations and communities. Many of these studies make untested assumptions about the various ecological processes that influence these distributions. The most common assumption, which is often unstated, is that distribution patterns reflect habitat selection by individual animals. However, very few studies have directly tested the relative importance of this process or considered other processes that may also affect the distribution of stream invertebrates. In this thesis I use a correlative survey in combination with manipulative field experiments and behavioural observations to investigate the factors and ecological processes that are associated with the distribution of hydropsychid caddisfly larvae in rocky upland streams in southeastern Australia.

No description available.

This study is concerned with the geographical distribution of the provision of nursery education in England and Wales from 1981 to 1994. It examines the extent to which the provision of nursery education is related to the need for the service. Davies's (1968) concept of territorial justice implies a positive correlation between need and provision. This interpretation of territorial justice is well established in the academic literature on the distribution of public services. Boyne and Powell (1991) have questioned whether a positive correlation between need and provision is always required for territorial justice. The key issue, it has been argued, is the requirement to consider the dimensions of need and service provision. This study assesses the validity of Boyne and Powell's (1991) analysis and builds substantially on its foundation A new set of criteria for the assessment of territorial justice is developed. These are presented as the alternative to the Davies (1968) criterion of territorial justice. Territorial justice is more appropriately concerned with 'equal provision for relevant dimensions of need'. The criteria for the evaluation of territorial justice developed in this study are applied to nursery education. The correlations provide a mixture of evidence of territorial justice and injustice in the quantity of service provision, together with territorial injustice in the quality of service. The Davies (1968) criterion of territorial justice would have indicated greater territorial justice in the provision of nursery education than that found using the new criteria. This study contributes at two levels to previous literature on territorial justice: a new method is used to assess territorial justice and new evidence on territorial justice within nursery education provision in England and Wales is presented. Further studies using the criteria developed in this study are required in other areas of public policy.

Models that are used for predicting species' potential distributions are important tools that have found applications in a number of areas of applied ecology. The majority of these models can be classified as correlative, as they rely on strong, often indirect, links between species distribution records and environmental predictor variables to make predictions. Correlative models are an alternative to more complex mechanistic models that attempt to simulate the mechanisms considered to underlie the observed correlations with environmental attributes. This study explores the influence of the type and quality of the data used to calibrate correlative models. In terms of data type, the most popular techniques in use are group discrimination techniques, those that use both presence and absence locality data to make predictions. However, for many organisms absence data are either not available or are considered to be unreliable. As the available range of profile techniques (those using presence only data) appeared to be limited, new profile techniques were investigated and evaluated. A new profile modelling technique based on fuzzy classification (the Fuzzy Envelope Model) was developed and implemented. A second profile technique based on Principal Components Analysis was implemented and evaluated. Based on quantitative model evaluation tests, both of these techniques performed well and show considerable promise. In terms of data quality, the effects on model performance of false absence records, the number of locality records (sample size) and the proportion of localities representing species presence (prevalence) in samples were investigated for logistic regression distribution models. Sample size and prevalence both had a significant effect on model performance. False absence records had a significant influence on model performance, which was affected by sample size. A quantitative comparison of the performance of selected profile models and group discrimination modelling techniques suggests that different techniques may be more successful for predicting distributions for particular species or types of organism than others. The results also suggest that several different model design! sample size combinations are capable of making predictions that will on average not differ significantly in performance for a particular species. A further quantitative comparison among modelling techniques suggests that correlative techniques can perform as well as simple mechanistic techniques for predicting potential distributions.

M.Sc. / DNA microsatellite primers were used to determine the population structure (genetic variation, heterozygosity, inbreeding, genetic distance and phylogenetic relationships) of Camelus dromedarius in southern Africa. A camel population from Sudan as well as an Alpaca population was included for comparison. The results obtained from the geographical distribution indicate that a number of small populations occur and that it can not be regarded as breeding units. These populations are isolated and widely distributed over South Africa, Namibia and Botswana. Information obtained from questionnaires indicates that nine true camel breeders exist in the southern African region. The genetic variation at seven loci of the southern African camel population (heterozygosity value = 0.604) is less than that of Sudan (0.680), an indication that inbreeding has occurred due to isolation. This was confirmed with the reduced number (five) of rare alleles occurring in the southern African population opposed to 30 rare alleles in the Sudan population. Both the southern African and Sudan camel populations have less genetic variation than the Alpaca population (heterozygosity = 0.757). Less genetic variation is present within the camel populations of southern Africa when results from more loci (12) are included (heterozygosity values range from 0.345 to 0.483). These values are comparable to those of endangered species such as wild dog (0.560) and cheetah (0.390). Low to moderate genetic differentiation was obtained between the three southern African populations (FST values range from 0.039 to 0.058). This was expected since the animals are derived from the same source. The results from this study will provide a scientific base for the execution of future studies. It is recommended that individuals with the identified rare alleles should be used in a breeding policy to prevent further inbreeding and to improve the current gene pool by increasing the genetic variation by either relocating or lending the individuals from and to the breeders or by making use of artificial insemination. / Prof. F.H. van der Bank

Forest birds located by their calls were related to mapping units of different land resource maps using a new method. The method involved computer programs which use mathematical descriptions of landform, forest canopy heights, and the nature of the bird's call to predict the area censused for each bird species from each listening station. Computer programs were also used to display locations made from each station, and to associate the locations and areas censused with different mapping units. Yellow-bellied Sapsucker (Sphyrapicus varius), Chestnut-backed Chickadee (Parus rufeseens), Steller's Jay (Cyanocitta stelleri), Winter Wren (Troglodytes troglodytes), Varied Thrush (Ixoreus naevius), and Swainson's Thrush (hylocichla ustulata) were each related to both serai stages and vegetation taxonomic units; while Hairy Woodpecker (Dendrpcopus villosus), Common Flicker (Colaptes cafer), Red-breasted Nuthatch (Sitta canadensis), and Olive-sided Flycatcher (Nuttallornis borealis) were each related only to seral stages. Most species showed a consistent pattern of selection for mapping types with repeated census. Results for the Steller's Jay indicated some change in the use of seral stages between census periods. However, there was no clear trend in use over time, and the observed changes may include effects of flocking which would violate the statistical assumption that locations were independent. Each species had a unique pattern of selection of seral stages and of vegetation types. Species with similar patterns of selection were grouped to form five groups for seral stages and three groups for vegetation types (groups not mutually exclusive). Only Chestnut-backed Chickadee with Yellow-bellied Sapsucker, Swainson's Thrush with Winter Wren, and Varied Thrush with Winter Wren were grouped together both for seral stages and for vegetation types. A more definite preference among seral stages than vegetation types was detected for most of the species studied. However, the Steller's Jay preferred only two of the vegetation types, while it used all seral stages somewhat equally. Most of the species studied preferred older seral stages. Common Flicker, Steller's Jay, and Olive-sided Flycatcher also used younger stages; while Swainson's Thrush selected for stages of medium age. Of the species related to vegetation types, only Chestnut-backed Chickadee did not show some preference for taxonomic units associated with high soil moisture. The preference was most pronounced for Steller's Jay, which concentrated its use on the two wettest types. The preference by Yellow-bellied Sapsucker may be explained by the older trees and snags that survived logging and fire in wet areas. All of the species also used many of the drier types. The data support the hypothesis that land resource maps can be used to predict the occurrence of wildlife. The results suggest that habitat for a wildlife species can be predicted over vast areas if the areas have been mapped, and if significant differences in the habitat value of different mapping units have been documented. The results indicate that the prediction can be improved by combining the predictions from two or more maps. The predicted area and spatial distribution of high-quality habitat can be compared with management policy objectives for the wildlife species. The predicted change in available habitat with planned forest management activities can provide criteria for habitat management. The same land resource maps may be used for many wildlife species, thereby facilitating multi-species habitat management. / Forestry, Faculty of / Graduate

This study of deer ecology was conducted from January, 1975, to May, 1976, in the East Kootenay region of British Columbia. Plant communities within the annual range of a population of white-tailed deer were classified into habitat types and analysed for species composition, productivity and forage utilization. Relative levels of use of winter habitat types were determined from track and pellet group counts. Use of summer habitats was determined subjectively from ground and aerial surveys. Movements associated with changes in habitat use and season were documented by observation and radio tracking of marked individuals. Forest succession over much of the winter range was found to have resulted in substantial decreases in availability of herbaceous and deciduous browse, causing quantitative and qualitative changes in the diet of this deer population. To compensate for this situation deer feeding activity was concentrated in the open habitat types which provide maximum quantities and qualities of forage. However, snowdepth in one winter was found to reach levels which prevented deer from exploiting these areas and concentrated them in areas where a maturing overstory reduced snowpack. The impact of concentration in shelter types with consequent reduction of available food, compared to a very mild winter, was an apparent 30% reduction in the juvenile:100 adults ratio the following spring. Summer distribution also appeared to be affected by large-scale forest succession which has produced a pattern of widely scattered, small openings. These were found mainly along water courses or rocky slopes at mid- elevations on the west side of the Rock Mountain Trench. Deer density was low throughout the summer range, but preference was observed for the open areas just described. Spring dispersal from the winter range was related to snow melt and green-up of vegetation, particularly cultivated alfalfa fields. Summer home ranges were relatively small and summer movements limited. The average distance travelled to summer range by nine deer was greater than that reported elsewhere in the literature and may be related to summer range habitat condition. Fall movements were apparently stimulated by lasting snowfall. Although density of deer on the winter range varied greatly between years, home range loyalty was found to be relatively high. / Science, Faculty of / Zoology, Department of / Graduate

No description available.