In certain soils, notably in those liable to crusting (p. 117), instead of heaving the soil, the water in freezing emerges bodily from small cracks, in foliated or wire-like forms (“ice-flowers”) resembling those of native silver, and formed substantially in the same way, by a kind of “wire-drawing” process, aided by crystallization. Small ice-crystals formed on the surface of small crevices filled with water cause others to be formed at their lower ends, and the expansion occurring in freezing, forces the ice upward; the process repeating itself under favorable conditions, until the stalks or sheets of ribbed ice grow to a height of several inches. This phenomenon is especially frequent in the middle cotton States—Arkansas, Tennessee, northern Mississippi, etc., where frequent changes from rainstorms or thaws to cold northwest winds occur in winter. CHAPTER VIII. SOIL AND SUBSOIL. CAUSES AND PROCESS OF DIFFERENTIATION. HUMUS. _Soil and Subsoil Ill-defined._—While the general mass of rock debris formed by the action of the agencies heretofore discussed as soil-material, may under proper conditions become soil capable of supporting useful plant growth, universal experience has long ago recognized and established the distinction between soil and subsoil: by which are ordinarily meant, respectively, the portion of the soil-material usually subjected to tillage, and what lies beneath. There can be no question about the practical importance of this distinction; but the definition of the two terms, as commonly given in some works of agriculture, is both incomplete and, in its application to many cases, partly misleading. The differentiation of soil and subsoil is due partly to the action of organic matter and micro-organisms, partly to physico-chemical causes, now to be discussed in detail. THE ORGANIC AND ORGANIZED CONSTITUENTS OF SOILS. _Humus in the Surface soil._—The most obvious mark of distinction between soil and subsoil is, usually, the darker tint of the former, due to the presence of humus or vegetable mold, which becomes most apparent by darkening of the tint when the soil is moistened. Thus soils having a gray tint when dry, may become almost black when wetted. When no such deepening of color occurs in wetting, the absence or great deficiency of humus may safely be inferred. The only other substance whose presence may invalidate the conclusions based upon the darkening of the soil tint, is ferric hydrate (iron rust), which itself possesses the property of darkening on wetting, and may effectually cover either the presence or the absence of humus.