Recent research has demonstrated promising results showing the possibility of separating arseniccopper sulphides from other copper minerals by controlling the potential of the flotation pulp. Most of these studies were conducted on single mineral systems, and the selective removal of arseniccopper minerals in real ore systems is not well understood. In real ore systems, the efficiency of the separation strongly depends on the mineralogical characteristics of the ore samples. This study seeks to understand the effect of ore mineralogy on the floatability of enargite in a complex ore system, under a controlled potential flotation environment.A composite of several high arsenic-containing drill core intersections for the high arsenic sample (HAS) and a composite of some low arsenic-containing drill core intersections for the low arsenic sample (LAS) were selected from the Tampakan copper-gold deposit in the Philippines, providing a range of arsenic levels. Arsenic in the HAS sample (enargite) was practically twice that for the LAS sample. The non-enargite copper minerals (NECu) were mostly chalcopyrite and bornite in both samples.Comprehensive size-by-size, chemical and mineralogical analyses were performed on both ore samples. It was observed that the two ore samples had similar mineralogical characteristics in terms of mineral content and liberation distribution, however there are some differences in the proportions of minerals. It was also observed that NECu minerals were mostly distributed to the coarser size fractions, while the proportion of enargite in the finer size fractions was higher than for NECu. The mineral grain size data showed that enargite had the finest grain size distribution compared to other copper minerals.The selective separation of enargite from NECu minerals in a rougher flotation system under controlled pulp potential was investigated for both samples. It was observed that it is possible to selectively separate enargite from other copper minerals after reducing the pulp potential to about -200 mV SHE at pH 11 in the LAS sample. However, no separation between enargite and NECu minerals was observed at a reducing potential for the HAS sample, and enargite did not float very well for this sample as the Eh was changed.The Particle Kinetic Model was used to predict the flotation response of enargite and NECu in the HAS sample based on mineral flotation rates derived from the LAS sample and the mineralogy of the HAS sample. It was observed that the predicted values of enargite recoveries were higher in the HAS sample when compared with the results of actual flotation tests. As the mineralogy of the two ore samples was similar in terms of mineralogical and liberation characteristics, the only reason for the iii poor prediction of the model for the HAS sample was due to the change in the flotation pulp conditions for the HAS sample. There were two possible reasons for this change in the flotation response of the HAS sample. One was the higher levels of pyrite present in the HAS sample com...