[en] The principle of using two gamma ray beams of different energies in the analysis of coal/stone mixtures is described. This technique has been developed in order to monitor run-of-mine coal on a conveyor belt and a prototype device has been built by Marconi Space and Defence Systems Ltd. A method of determining the stone content of a coal/stone mixture consists of measuring the uncollided flux transmitted through the mixture from two gamma ray sources of different energy. As only the uncollided flux is measured, the method is independent of the spatial distribution of the two components in the mixture. For best results, the gamma-ray energies should be chosen such that at one of them the absorption coefficients of the coal and stone are almost equal while at the other, there should be a substantial difference between them. Choosing the energies in this way means that one of the expressions for the mass loading will exhibit only a small dependence on the stone content. It should also be borne in mind that low energy radiation is easily absorbed by minerals and this could limit the maximum mineral loading which can be usefully analysed. At the other end of the energy scale, gamma-ray beams are less readily absorbed in passing through large amounts of mineral and hence the sensitivity to changes in mass loading will be poor. One choice of energies which satisfies all of these constraints is 60keV radiation from Am₂₄₁ and 662 keV radiation from Cs₁₃₇. A schematic diagram of a prototype device operating on the above principle is shown in Figure 1. The problems encountered during the development of this instrument are described under the following headings: downscatter and background interference, randomly varying loads and counting losses due to pulse pile-up