Soil physical, chemical, and taxonomic data of about 12,000 pedons from the continental U.S., Hawaii, Puerto Rico, and some foreign countries were used to develop multiple regression equations to estimate sum of exchangeable bases (SUMBAS) and base saturation (BS). Soils were grouped according to their taxonomic classification at order and suborder categories. Multiple regression equations using organic carbon and clay contents, and soil pH in water ratio 1:1 accounted for more than 50% of the variation in SUMBAS in 11 of 35 suborders included in this study. Regression equations using organic carbon and clay contents, pH, 1M KC1 extractable Al or percent Al saturation, and cation exchange capacity accounted for more than 70% of the variation in SUMBAS in 18 of 35 suborders. Percent Al saturation and organic carbon content were negatively related to BS. These two parameters accounted for more than 70% of the variation in BS in 16 of 32 suborders. Regression equations using soil pH alone, in turn, accounted for more than 70% of the variation in BS in 4 of 6 suborders with a formative element Aquic in their taxonomic names.