The flotation response of gersdorffite has been determined in a series of batch flotation tests in which a sulphide sample containing about 76% gersdorffite was floated from mixtures with quartz. Using xanthate as collector, the arsenide floated strongly at pH 9, irrespective of whether lime, sodium hydroxide or sodium carbonate was used as pH modifier. When the pH was raised further to 10 or more using lime, the flotability decreased sharply. Changing the pulp potential at pH 9 also strongly affected the flotation response. Above about -230 mV SHE, the mineral floated strongly, but below this potential it was essentially non-flotable. As far as it was tested (up to +400 mV SHE), no upper limiting potential was found. Adding cyanide to the grinding mill brought about only a weak depressant effect on subsequent gersdorffite flotation at both pH 9 and 10. However, adding the same amount of cyanide to the flotation cell produced a strong depressant effect at the same pH values. At both pH 9 and pH 10, a threshold cyanide addition existed below which gersdorffite floated strongly, and above which it did not. For the conditions used in this study, this addition equates to about 100 g/t at pH 9 and about 80 g/t at pH 10. A hypothesis for the depression effect is presented which is consistent with both the available flotation data and a series of diagnostic analyses of grinding and flotation pulps completed during the test work. Crown Copyright (C) 2009 Published by Elsevier B.V. All rights reserved.