Clay mineral quantification by XRD is difficult when mixed-layer clay minerals and discrete clay types are both present. New procedures for peak decomposition and pattern simulation offer increased opportunities to obtain mineral abundance estimates. This proposed methodological sequence, for quantitative representation (QR) of complex clay samples, involves: (1) determination of layer type, mixed-layer proportion and order (R); (2) simulation of XRD patterns using MULCALC, an adaptation of NEWMOD; and (3) interpretation of the clay assemblage by fitting the whole pattern with CLAY++, a statistical program. The product is a QR of individual phases or a summation of layer types. The absence of quantitative reference standards means results cannot be checked for accuracy, but the statistical fit is highly reproducible and less prone to operator error. The QRs may be obtained with simulated or actual reference mineral patterns in the database. Results for freshwater marsh samples illustrate the approach.