As remaining gold deposits tend to become more refractory, some form of oxidative pre-treatment involving hydro or pyrometallurgy is necessary to maintain high metal recoveries. These pre-treatments include pressure oxidation, bio-leaching or roasting. In the past, the primary criterion for selection of oxidative pre-treatment relied mainly on the extent of gold recovery and relative economics. In this study, a different approach has been adopted, ie the quantitative determination of environmental performance of the oxidative pre-treatment processes bio-leaching, roasting and pressure oxidation to assess which had the least impact on the environment. By establishing the environmental performance of processes, waste minimisation can be addressed early in process selection. Using mass balances, the environmental theme method was applied to determine the performance, on the basis of gold treated, of the oxidative pre-treatment processes of selected plants. No clear trend was observed in the overall impacts of each of seven plants, with all but one of the environmental evaluation indices being of the same order of magnitude; this was probably due to masking effects arising from the large variations in feed composition, tonnage, and gold levels. Within each type of pre-oxidation plant however, certain trends were detectable. For instance, irrespective of plant size, ore composition or mineralogy, for the bio-leaching and pressure oxidation units, the general order of environmental impact from greatest to least was sequentially toxicity, acidification potential, depletion of resources, solid waste generation and greenhouse enhancement. For the roasters the acidification potential was the most significant. Determination of the environmental performance of the oxidative pre-treatment of a hypothetical ore in a plant of fixed tonnage was undertaken to remove the possible masking effects of variables associated with feeds of different mineralogy, gold tenor and tonnage. The bio-leaching process gave a marginally better performance than pressure oxidation, the roasting process the worst, although there was not a great difference between them. The decreasing order of contribution to the environmental index was acidification potential, toxicity, resource depletion, solid waste generation and greenhouse enhancement for bio-leaching and pressure oxidation; whereas it was acidification potential, solid waste generation, toxicity, resource depletion and greenhouse enhancement for roasting. Since the major contributor to the environmental index was the acidification potential, the environmental performance of treatment processes for the acidic effluents was calculated. After neutralisation of the net acid resulting from bio-oxidation and pressure oxidation, and the application of various scrubbing and product formation processes for the sulphur dioxide from the roasting process, it was shown that pressure oxidation gave a marginally better performance than bio-leaching, ahead of roasting. This study demonstrated the potential for the development of a predictive tool which could be used in determining the environmental performance, and thus the selection, optimisation, and improvement of pre-oxidation processes for least impact on the environment. These would allow operation of these processes within defined environmental performance limits and decisions regarding the desirability of processing various parts of a heterogeneous orebody to be made on environmental grounds.