Abstract The mineral referred to as kaolinite/smectite has been identified in soils and paleosols from Holocene to Pennsylvanian in age (and perhaps older) and most fireclays, ball clays, and other poorly crystallized kaolinites. It is a mixed-layered mineral composed of an expandable 2:1 (9.7Å, collapsed) layer and a kaolinitic 1:1 (7Å) layer. Models of bonding between layers or units of this mineral can be calculated by using stacking sequences of three structural modules-2:l layers of parent material, 1:1 layers, and 2:1 layers hydrogen bonded to a 1:1 layer. This structural modeling suggests constraints on mechanisms of kaolinitization and types of interlayers that will expand, bond, form stable crystallites, or some combination of these. It also suggests possible ordering mechanisms and limits to particle size. Layer charge on the 2:1 parent material and on the 2:1 and 1:1 alteration products ranges from the maximum of that for illite to a minimum near zero. The wide range of possible layer charge on the 2:1 layer suggests that mixed-layered kaolinite/expandables ( K/E ) would be a better name than kaolinite/smectite. The mechanism of formation of K/E is poorly understood. Our evidence supports a mechanism in which tetrahedral sheets are stripped from 2:1 layers and bonding occurs between the hydroxyl sheet of the newly formed 1:1 layer and the adjacent 2:1 layer. Inherited octahedral and tetrahedral substitutions may result in atypical 7 Å layers. Growth of K/E may be terminated by encounters between crystallites with opposite c* directions of their 1:1 layers or by strains resulting from inherited substitutions within the 7 Å layer. Pedogenic K/E occurrences are correlated with iron substitution within kaolinitic layers of K/E. Where order can be determined, we have observed ~R1.5. R1 has been reported in the literature. There appear to be two continuous genetic sequences within the kaolin group: 1) a series from allophane through halloysite, and 2) a series from 2:1 parent materials through K/E . Transformations from halloysite or K/E to well-crystallized kaolinite probably require recrystallization and therefore the last step in both sequences is discontinuous. Conceptually and structurally, we can make several useful comparisons between smectite to illite or illite to smectite transformations and the 2:1 to K/E transformation. Detection of kaolinite/expandibles is readily made by XRD studies of < 2 µ m or finer fraction samples after air drying, ethylene glycol solvation, and a heating routine (300°C, 350°C, 400°C). K/E with a composition near kaolinite-a peak near 7Å--can be distinguished from halloysite and well-crystallized kaolinite by the rapid intercalation of the latter two phases by many agents. This intercalation shifts the kaolinite and halloysite peaks to the 10Å to 12Å area of the diffractogram and leaves only the peak for K/E in the area between 7Å and 10Å. The 17Å peak of expandable-rich K/E (001 kaol /001 exp ) with ethylene glycol solvation is extremely broad even at low percentages of 7Å interlayering. This peak broadening distinguishes K/E with a low proportion of kaolinitic layers from smectite and I/S peaks near 17Å. After heating to 300°C, elevated background intensity or a peak on XRD traces between the 7Å and 10Å positions is the most sensitive diagnostic method to detect K/E. Loss of the high-angle, K/E shoulder on the 10Å peak (assuming one or more discrete 2:1 phases are present) after heating to 350-400°C can be an equally sensitive method for detecting and quantifying K/E. The difficulty of detecting K/E, especially samples with low contents of kaolinitic interlayers, suggests that K/E is much more widespread than previously thought. The A and B zones of a soil profile typically contain the most K/E and the highest proportion of kaolinitic to expandable layers. The parent 2:1 clay minerals, climate, plant community, and degree of drainage determine whether K/E, halloysite, well-crystallized kaolinite, bauxite, or a combination of these phases forms. Determination of the types and amounts of kaolin minerals in soils may offer valuable insights into the nature of soil formation and associated processes and rate of soil formation below unconformities.