Flash flotation cells are increasingly being employed to recover valuable material present in the grinding circuit, reducing the potential for over-grinding and enhancing plant performance with the added advantage of reduced capital outlay and operating costs. However it is not always possible to quantify the contribution of the flash to overall recovery or to predict how this will alter with changing feed ore properties. This paper is the second in a series of articles written on the topic of coarse particle and flash flotation and examines the nature of the particles being recovered by an industrial flash cell and compares them with those recovered by a laboratory batch flotation cell. The applicability of batch flotation test methods for predicting whether an ore is amenable to the flash flotation process is investigated and shows that a targeted batch flotation test can be used in conjunction with mineralogical analysis to predict the response of the target (valuable) mineral to an industrial flash flotation process. Laboratory tests were conducted on a refractory gold ore taken as a belt cut, allowing the results of the laboratory analysis to be directly compared with the actual plant performance of the same ore. Direct comparison of laboratory flotation test concentrates with that of the plant flash flotation cell shows that the mineralogical response observed in a batch flotation test can be used to predict the nature of the particles that will be recovered in an industrial flash flotation cell. Both the size distribution of the concentrate and upgrade ratios of the coarse size fractions are able to be determined by laboratory methods; however differences in the recoveries and kinetic parameters between laboratory and plant were observed. This paper discusses these results and demonstrates how the tests can be used in conjunction with mineralogy data to predict the amenability of an ore to the flash flotation process. (C) 2012 Elsevier Ltd. All rights reserved.