Dilution in underground coal operations can be underestimated by up to five times during feasibility studies, with disastrous economic consequences. Every one per cent of stone waste introduced can reduce an in situ specific energy of 25 MJ/kg by some 0.25 MJ/kg and increases the abrasiveness of the run-of-mine coal. Where a wash plant is not in place, penalties imposed for not meeting contractual specifications alone can amount to some $A 1 million per year. This paper presents a method for monitoring, reporting and ultimately controlling dilution in an underground coal mining operation and presents the dilution trends from a continuous miner/longwall mine over an eight year period. The overall dilution level for the mine can be back-calculated by comparing the run-of-mine coal quality to the predicted in situ, undiluted quality. However, this will not identify the actual contaminant types, areas of the mine responsible for the dilution and the reasons for the dilution. This additional information is essential for controlling dilution to acceptable levels. Quality and horizon control using existing sampling information is generally only adequate for longer-term control purposes, and is not suitable for monitoring mining-induced dilution. The cost of close-spaced surface drilling to collect the information necessary for short-term control is prohibitive, particularly for deeper mines. However, much of this data can be collected through underground channel sampling and horizon control measurements taken during continuous miner development and longwall extraction. The mine runs the risk of not meeting contractual quality specifications and of increased equipment wear if this additional information is not available and the sources of dilution cannot be controlled. The information can also be used to plan longwall extraction heights on a short-term scale with resulting improvements in productivity.