The word “soil,” like many common words, has several meanings. In its traditional meaning, soil is the natural medium for the growth of land plants, whether or not it has discernible soil horizons. This meaning is still the common understanding of the word, and the greatest interest in soil is centered on this meaning. People consider soil important because it supports plants that supply food, fibers, drugs, and other wants of humans and because it filters water and recycles wastes. Soil covers the earth’s surface as a continuum, except on bare rock, in areas of perpetual frost or deep water, or on the bare ice of glaciers. In this sense, soil has a thickness that is determined by the rooting depth of plants.

        About 1870, a new concept of soil was introduced by the Russian school led by Dokuchaiev (Glinka, 1927). Soils were conceived to be independent natural bodies, each with a unique morphology resulting from a unique combination of climate, living matter, earthy parent materials, relief, and age of landforms. The morphology of each soil, as expressed by a vertical section through the differing horizons, reflects the combined effects of the particular set of genetic factors responsible for its development.

        This was a revolutionary concept. One did not need to depend wholly on inferences from the underlying rocks, the climate, or other environmental factors, considered singly or collectively; rather, the soil scientist could go directly to the soil itself and see the integrated expression of all these in its morphology. This concept made it not only possible but also necessary to consider all soil characteristics collectively, in terms of a complete, integrated, natural body, rather than individually. Thus, the effect of any one characteristic or a difference in any one depends on the others in the combination. Experience has shown that no useful generalizations about single characteristics can be made for all soils. Characteristics are given weight according to the knowledge gained through research and experience in soil genesis and the responses of soil to management or manipulation. Both research in genesis and the responses of soils have vital roles, but they are themselves one step removed from the taxonomy of the soil, which is based on combinations of soil characteristics. In short, the new concept made pedology possible.

        The Russian view of soils as independent natural bodies that have genetic horizons led to a concept of soil as the part of the earth’s crust that has properties reflecting the effects of local and regional soil-forming agents. The solum in that concept is the set of genetic horizons developed by soil-building forces, but the parent material beneath is nonsoil. This concept has limitations. If a solum is 1 or 2 m thick, there is little conflict between the concept of soil as solum and the concept of soil as the natural medium for the growth of terrestrial plants. If genetic horizons are thin or absent and unconsolidated parent material lies at or only a few centimeters below the surface, there is serious conflict between the concepts. Dokuchaiev realized this conflict and, despite the lack of horizons, included young alluvium and peat in his classification of soil.

        Soil in this text is a natural body comprised of solids (minerals and organic matter), liquid, and gases that occurs on the land surface, occupies space, and is characterized by one or both of the following: horizons, or layers, that are distinguishable from the initial material as a result of additions, losses, transfers, and transformations of energy and matter or the ability to support rooted plants in a natural environment. This definition is expanded from the previous version of Soil Taxonomy to include soils in areas of Antarctica where pedogenesis occurs but where the climate is too harsh to support the higher plant forms.

        The upper limit of soil is the boundary between soil and air, shallow water, live plants, or plant materials that have not begun to decompose. Areas are not considered to have soil if the surface is permanently covered by water too deep (typically more than 2.5 m) for the growth of rooted plants. The horizontal boundaries of soil are areas where the soil grades to deep water, barren areas, rock, or ice. In some places the separation between soil and nonsoil is so gradual that clear distinctions cannot be made.

        The lower boundary that separates soil from the nonsoil underneath is most difficult to define. Soil consists of the horizons near the earth’s surface that, in contrast to the underlying parent material, have been altered by the interactions of climate, relief, and living organisms over time. Commonly, soil grades at its lower boundary to hard rock or to earthy materials virtually devoid of animals, roots, or other marks of biological activity. The lowest depth of biological activity, however, is difficult to discern and is often gradual. For purposes of classification, the lower boundary of soil is arbitrarily set at 200 cm. In soils where either biological activity or current pedogenic processes extend to depths greater than 200 cm, the lower limit of the soil for classification purposes is still 200 cm. In some instances the more weakly cemented bedrocks (paralithic materials, defined later) have been described and used to differentiate soil series (series control section, defined later), even though the paralithic materials below a paralithic contact are not considered soil in the true sense. In areas where soil has thin cemented horizons that are impermeable to roots, the soil extends as deep as the deepest cemented horizon, but not below 200 cm. For certain management goals, layers deeper than the lower boundary of the soil that is classified (200 cm) must also be described if they affect the content and movement of water and air or other interpretative concerns.

        In the humid tropics, earthy materials may extend to a depth of many meters with no obvious changes below the upper 1 or 2 m, except for an occasional stone line. In many wet soils, gleyed soil material may begin a few centimeters below the surface and, in some areas, continue down for several meters apparently unchanged with increasing depth. The latter condition can arise through the gradual filling of a wet basin in which the A horizon is gradually added to the surface and becomes gleyed beneath. Finally, the A horizon rests on a thick mass of gleyed material that may be relatively uniform. In both of these situations, there is no alternative but to set the lower limit of soil at the arbitrary limit of 200 cm.

        Soil, as defined in this text, does not need to have discernible horizons, although the presence or absence of horizons and their nature are of extreme importance in soil classification. Plants can be grown under glass in pots filled with earthy materials, such as peat or sand, or even in water. Under proper  conditions all these media are productive for plants, but they are nonsoil here in the sense that they cannot be classified in the same system that is used for the soils of a survey area, county, or even nation. Plants even grow on trees, but trees are regarded as nonsoil.

        Soil has many properties that fluctuate with the seasons. It may be alternately cold and warm or dry and moist. Biological activity is slowed or stopped if the soil becomes too cold or too dry. The soil receives flushes of organic matter when leaves fall or grasses die. Soil is not static. The pH, soluble salts, amount of organic matter and carbon-nitrogen ratio, numbers of microorganisms, soil fauna, temperature, and moisture all change with the seasons as well as with more extended periods of time. Soil must be viewed from both the short-term and long-term perspective.