Comparisons are presented between the results of quantitative X-ray diffractometry using a Rietveld-based computational method (SIROQUANT) on low-temperature ash (LTA) and normative calculation from chemical analysis of high-temperature ash for a series of coal samples from the Callide Basin of Queensland. Mineral percentages indicated by each method are also compared to the results of conventional X-ray diffractometry on low-temperature ash spiked with a known proportion of corundum, evaluated in conjunction with detailed clay mineral analysis of the < 2 mu m fraction of each sample by oriented-aggregate X-ray diffraction methods. For species where direct comparison is possible, the proportions of the different minerals in the crystalline fraction indicated by SIROQUANT are similar to those indicated by normative analysis using a computational procedure referred to as SEDNORM. More conventional XRD study of spiked LTA materials, with the samples studied, seems to underestimate quartz and siderite content, and by default overestimate kaolinite. SIROQUANT, on the other hand, tends to overestimate the proportion of Ca-bearing minerals, relative to the Ca content of the high-temperature ash, processed using the SEDNORM computation. Both the SIROQUANT analysis and the normative evaluation of SEDNORM provide significantly more quantitative information on coal mineralogy than other more commonly used techniques.