Endocrine disrupting chemicals (EDCs) are ubiquitous in the environment and have been detected in drinking water. Although various water treatment processes can remove EDCs, chemicals can migrate from pipes that transport water and contaminate drinking water. Globally bottled water consumption is steadily rising as an alternative to tap water, but EDCs have also been detected in bottled water. Sources of EDCs in bottled water include contamination of the water source, contamination through the production process or the migration of EDCs from the packaging material. There is limited information on EDCs in drinking water and bottled water from South Africa. The aim of this study was to determine the estrogenic activity, levels of selected EDCs and the potential health risks associated with the consumption of water from selected distribution points in Pretoria (City of Tshwane) and Cape Town as well as bottled water. The study consisted of 3 phases. Phase 1 included the analysis of drinking water samples from ten water distribution points in Pretoria and Cape Town collected over four sampling periods. In phase 2, ten brands of bottled water were analysed after exposure to different storage conditions (20°C, 40°C, light and dark) for 10 days. Samples were extracted using a C18 solid phase extraction method. Estrogenic activity was assessed using the recombinant yeast estrogen screen (YES) bioassay and the T47D-KBluc reporter gene bioassay. The extracts were analysed for di(2-ethylhexyl) adipate (DEHA), di(2-ethylhexyl) phthalate (DEHP), diisononylphthalate (DINP), dibutyl phthalate (DBP), bisphenol A (BPA), nonylphenol (NP), 17β-estradiol (E2), estrone (E1) and ethynylestradiol (EE2) using UPLC-MS. Phase 3 consisted of a scenario based health risk assessment to assess the carcinogenic and toxic human health risks associated with the consumption of distribution point and bottled water. All the samples were below the detection limit (dl) in the YES bioassay, but estrogenic activity was detected in bottled and distribution point water using the T47D-KBluc bioassay. All samples were below the 0.7 ng/L trigger value for estrogenic activity in drinking water. NP was below the dl for all the samples, E2 was detected in five distribution point samples and E1, EE2, DEHA, DEHP, DINP, DBP and BPA were detected in distribution point and bottled water samples. The estrogenic activity and levels of target chemicals were comparable to the levels found in other countries. Hazard quotients for BPA, DEHA and DINP were higher in bottled water compared to distribution point water. The greatest non-carcinogenic health risk was posed by E1 in distribution point water from Pretoria and the highest cancer risk by levels of DEHP in distribution point water from Cape Town. However, overall, health risk assessment revealed acceptable health and carcinogenic risks associated with the consumption of distribution point and bottled water. Although the potential health risks posed by the EDCs found in the water samples in this study were low, the fact that potential EDCs were found in the water samples are still of concern. A monitoring strategy that also includes water from other municipalities and other brands of bottled water are therefore recommended. / Thesis (PhD)--University of Pretoria, 2016. / School of Health Systems and Public Health (SHSPH) / PhD / Unrestricted