Desert soils have less water, are non-flushing, and have less chemical weathering, more carbonate, and soluble minerals than humid soils. Still, authigenic silicate clays and argillic horizons can form. The mechanisms of argillic horizon formation, its destruction, and the sources of clay minerals were important questions for the Desert Soil-Geomorphology Project in New Mexico during the 1950s to 1970s because of the need to understand basic principles of soil genesis and soil taxonomy in desert regions. Mica, kaolinite, and smectite are most abundant in Desert Project soils with mica and kaolinite changing little with depth, age, and parent materials, in contrast to smectite that increases with age and in B horizons. Less common are palygorskite, sepiolite, chlorite, and mixed-layer clay minerals. Chronologically, argillic horizons first form in Holocene soils with igneous parent materials. Limestone parent materials generally lack argillic horizons. Clay skins are on peds are most prominent in pipes of Bt material that penetrate calcic horizons.   The obliteration of clay skins result from engulfment by carbonates, landscape dissection, and termites, ant, and rodent burrowing.  Desert Project soils contain a mixture of inherited and neoformed clays. Mica, kaolinite, and mixed-layer clays are largely derived from dust and sedimentary parent materials, including older soils. Kaolinite, however, can also be neoformed along with smectite and vermiculite in monzonite residuum of mountain slopes and pediments. Palygorskite and sepiolite are neoformed in petrocalcic horizons, possibly as the result of pressure-solution generated by progressive amounts of calcite precipitation.