The objective of this investigation was to determine the elemental composition and mineralogy of the smectitic and illitic phases in interstratified smectite-illite from a typical upper Midwestern agricultural soil. A sample of soil clay ( < 2-mu-m fraction) from the Ap horizon of a Webster (fine-loamy, mixed, mesic Typic Haplaquoll) pedon was fractionated by centrifugation to prepare various fine-clay (< 0.020-, < 0.026-, < 0.036-, < 0.045-, < 0.060-, and < 0.090-mu-m size fractions) samples. X-ray diffraction analyses indicated that interstratified smectite-illite was the only mineral present in significant quantities in the fine-clay samples. Elemental analyses of the fine-clay samples were performed by inductively coupled plasma-atomic emission spectroscopy using suspension nebulization. The elemental analyses revealed significant linear correlations for the K vs. other element relationships (P < 0.001 for K vs. Si, Al, Fe, Ca, Ti, Mn, and Zn, and P < 0.01 for K vs. Mg). A nonlinear parameter optimization technique, based on elemental mass balance, was used to estimate the elemental compositions and proportions of the smectitic (K-free) and illitic (K-bearing) phases in the fine-clay samples. The results for the elemental mass balance optimization were independently verified by water mass balance. The smectitic phase was a high-charge (0.482 per formula unit), Fe-rich montmorillonite (46.6% tetrahedral charge). The illitic phase consisted of dioctahedral, tetrahedrally charged (86.9% tetrahedral charge) elementary illite particles. Layer charge for the illitic phase was 0.473 per formula unit, which is substantially below the lower limit for classification of illite. The interstratified smectite-illite comprised at least 60% of the whole (< 2-mu-m fraction) soil clay and the illitic phase comprised at least 36% of the interstratified smectite-illite.