Clay minerals are key components of soils. They may adsorb and release cations essential for plant nutrition (potassium, calcium, ammonium…). Recent improvements in X-ray data treatment allow a better quantification of soil clay minerals. In particular, peak decomposition programs associated with Reynolds model permit one to estimate total illite, smectite and hydroxy-interlayered quantities from a complex X-ray diffractometer. This new tool is of great interest in the quantification of clay mineralogy differences in soil sequences. Old studies and some new data reveal that the vegetation and land-use influence greatly micaceous clay minerals in soils. Illite layer increase near the surface seems to be a general trend in prairie and deciduous forest soils. On the contrary, intensive agriculture without potassic fertilization leads to a quick decrease of illite layer. We hypothesize that these modifications are strongly linked with the potassium cycle in soils. The increase of illite layers near the surface conforms with the “nutrient uplift hypothesis” expressed by Jobbagy and Jackson where some elements are selectively moved up to the top of a soil profile by plants. In particular, uplifted potassium may be fixed by clay minerals and promote illite layers formation. Conversely, potassium depletion induced by intensive agriculture may lead to illite destabilization. An order of magnitude of the potassium involved in these transfers was estimated thanks to laboratory experiments. Indeed, our experiments showed that there is a linear relationship between the potassium content and illite quantity in clay minerals. Potassium quantity in soils corresponding to illite layer modification can therefore be estimated. It appears that roughly 7 years of prairie plant growth are necessary to uplift enough potassium to sustain one year of intensive corn cropping.