Agriculture plays a significant role in South Africa’s export earning and in providing employment opportunities. Amongst the major agricultural crops is citrus. Within the context of postmodern and feminist geographies and utilizing the intensive research design, this thesis discusses citrus production with a focus on growers and black women workers in the Kat River Valley area in Fort Beaufort, Eastern Cape. The thesis also explores the challenges and implications of restructuring within agriculture for growers, cooperatives and labour within the citrus industry. It is argued that whilst legislation related to labour and tenure within agriculture has changed since the mid-1990s, power relations in the citrus industry have remained firmly entrenched. Consequently, the flexible labour strategies that the citrus farmers adopted have had serious consequences for women worker

Livelihood stressors in southern Africa, such as HIV/Aids and climate change, do not act in isolation but rather interact concurrently in complex socio-ecological systems with diverse, interrelated and compounded affects. Households experience differential vulnerability to such stressors based on contextual factors such as geographical location, income level and the gender and age of its members. Households’ differential experiences of vulnerability are further defined by the households’ use of their capital stocks: the human, social, natural, financial and physical capital available to the household to form livelihoods and resist the detrimental effects of a stressor. The capital stocks of 340 households were measured in two sites in the Eastern Cape, South Africa, using a household survey. These data were analysed to determine differences between the sites, households with heads of different gender and households of different income levels. Further data relating to the drivers and interactions of stressors over temporal and spatial scales, as well as the perceived value of various forms of capital by different social groups in the two sites, were collected via Participatory Learning and Action (PLA) methods including timelines, mental modelling and pair-wise ranking. Although the two sites have similar levels of income and fall within the same province, many significant differences emerged. The two sites showed different distributions of household head genders and different stressors and perceptions of vulnerability, perhaps owing to differences in their capital stocks, acting alongside the influence of culture and access on a shifting rural-urban continuum. These discrepancies further transpired to reflect crucial differential experiences along gender lines and income levels in each site. Vulnerability was often context specific, not only because of unique drivers of stress in different areas, but also because socio-economic groups and localities often had characteristics that could potentially exacerbate vulnerability, as well as characteristics that can potentially facilitate adaptive capacity. Stressors were found to have depleted multiple forms of capital over time, while new stressors were emerging, raising concerns over the most appropriate means of social protection within these contexts.