Dispersive soil has poor physical properties, which reduce porosity, air and water movement. These properties can reduce crop production and create problems with soil stability under buildings and infrastructure. Traditionally, clay dispersion has been related to the exchangeable sodium percentage (ESP) of soil or to the sodium adsorption ratio (SAR) of soil solutions. However, the roles of the other exchangeable cations, potassium and magnesium, in clay dispersion have been debated in the literature and their importance is unresolved. Based on the relative dispersive and flocculating powers of sodium, potassium, magnesium and calcium, a new concept for describing dispersive soil is proposed. This defines the net dispersive charge as the difference between the dispersive charge, calculated from the concentration of exchangeable cations measured at a given soil pH and their dispersive powers, and the flocculating charge, calculated from the concentrations of cations in dispersed solutions and their flocculating powers. The net dispersive charge of a soil determines the clay dispersion. Clay disperses when the net dispersive charge is positive and does not disperse when it is zero or negative. This new concept resolves controversies in the literature caused by the roles of organic matter, clay mineralogy, exchangeable cation composition and electrolyte concentration in relation to clay dispersion. Experimental confirmation of this concept is provided by the analytical data of 96 soil samples collected from different parts of South Australia and Victoria in Australia and from previously published data. Possible methods of estimating net dispersive charge from soil solutions are discussed.