Data in the literature for soils that are dominated by each of the main types of clay minerals were examined and compared with those for reference clay minerals of the same types to determine the extent to which the nature and properties of clay-size minerals in soils could be explained by those of clay minerals with the same name from non-soil, ‘geological’ environments. Published information on soils from Australia, New Zealand and Iran was sourced for this study. The clay fractions of each of the soils are dominated by either one of the common phyllosilicates: kaolinite, halloysite, illite/mica, vermiculite, smectite, and palygorskite, or by the nanocrystalline mineral, allophane. Data for samples of kaolinite that had been extracted from soils from several countries (Australia, Thailand, Indonesia and Brazil) and purified before characterization have also been examined. In soils, each dominant clay mineral is generally associated with other materials, including iron oxides, other phyllosilicates and/or nanocrystalline minerals and organic matter. As the most studied example of an extracted phyllosilicate, kaolinite shows a wide range of properties in different soils, but a narrower range of properties within a particular locality. However, almost all of the soil kaolinites studied have larger specific surface areas and higher cation exchange capacities than reference kaolinites. The literature also reveals that, among phyllosilicates in soils, illites have a wide range of potassium contents, expandable minerals (vermiculites and smectites) may be interlayered by hydroxy-Al species particularly, and smectitic layers often occur interstratified with other layers, including those of illite, kaolinite and halloysite. The variability of soil phyllosilicates and their common association with other, often poorly crystallized but highly reactive minerals and compounds can be explained by their formation in the highly heterogeneous and dynamic soil environment. Phyllosilicates from non-soil or geological sources are poor models for the representation of secondary clay-size minerals in soils. In philosophical terms, the reduction of soil mineralogy to mineralogy as it is practiced within geology is misleading because of the differences between the minerals formed in soil and geological environments. In other words, clay minerals as they are defined as mineralogical entities for geology are of a different ‘kind’ to clay minerals in soils and cannot serve as ‘types’ or ‘stereotypes’ to enable explanation of the contribution of secondary clay-size minerals to soil properties or behavior. It is more useful to view clay minerals in soils as secondary inorganic compounds of clay-size than to follow their definition for non-soil purposes as plastic phyllosilicate minerals.