|dc.description.abstract||Human activities have dramatically altered the functioning of ecosystems through the ages. Urbanization illustrates the effect of anthropogenic activity by the transformation of natural areas to ecologically disturbed regions (development of towns, cities and settlements). The growing need for urban employment in South Africa has led to an increase in the number of informal settlements on the periphery of urban areas. These settlements result in fragmentation and sprawling of cities, which intensifies strain on the natural environment. Fragmentation in urban regions then leads to the formation of 'patches' of land which exhibit different disturbance levels and are generally typified as either urban, suburban or rural areas. These land use types may be ecologically studied along an urbanization gradient, with the intention of obtaining meaningful comparisons. An urbanization gradient contains an urban landscape which consists of a densely built and developed core surrounded by an area of decreasing development and increasing 'naturalness'. The use of urbanization gradients has been proven world-wide as a useful tool for the study of changes in ecological patterns and processes across landscapes. This approach has been used to examine many different impacts of urbanization, namely on invertebrate communities, bird community composition and plant community composition. Using biological indicators to determine the degree of anthropogenic impact on the environment has proven effective in past studies. These indicators can be used to monitor ecological change following habitat disturbance, identify changing trends over time, provide early warning systems of degradation and diagnose the cause of existing problems. Several authors have supported the use of arthropods as suitable indicators of disturbance. The aim of this study was to determine what impact disturbance, due to urbanization, may have had on the diversity and abundances of epigeal (surface roaming) arthropods (focussing on ants, beetles and spiders) following an urbanization gradient approach. In addition, plant and soil data were combined with the arthropod analysis for each site studied, in order to obtain a better picture of how arthropod community composition would change in relation to these factors. The ant group were the numerically dominant group of the arthropods studied, although the beetles did have the highest number of species captured. Spiders were caught in low abundances, but were also represented by a high number of species. Dramatic decreasing trends were observed with respect to the ant abundances and diversity from rural to the more urbanized sites. Quite the opposite, seemed to occur with the beetles and spiders, who were dominant in species and numbers in the urbanized areas. This trend may be explained on account of the occurrence of generalists and opportunistic beetle and spider species, which seem to thrive in these heterogenous urban habitats.
When considering environmental components, percentage bare-ground and sand concentration seemed to be the determining factors in the rural sites, around which the ant group aggregated. Sandy habitats with patches of bare-ground provide more favourable micro-habitats for the ant species to roam and scavenge in, and are advantageous for nest building. Clay concentration and abundance of fruit seemed to assist in providing favourable habitats for the opportunistic and generalist beetle species, in the urbanized areas. High clay concentrations in the urban areas provided ideal conditions for abundant organic covering which would favour saprophagous (feed on decaying organic matter) beetle species and support diverse prey for the predatory beetle and spider species to feed on. Abundance of fruit may have attracted numerous herbivorous beetles (frugivorous beetles). Urbanization seemed to have a more pronounced effect on ant diversity and abundances in comparison to the beetles and spiders, and therefore recommended for future utilization as a suitable "Bio-indicator" group for further local disturbance studies.||