Test methods for evaluating the durability of a metal in soil are described in this article. Specimen design, preparation, burial, and retrieval techniques are discussed. The type of information sought during soil-induced corrosion evaluation controls the design configuration and the nature of the corrosion measurements. Consideration of these factors during the planning stage will help the corrosion engineer obtain a maximum amount of information with a minimum number of problems. The corrosion of metals in soils, which is also called underground corrosion, can be divided into two broad categories: corrosion in undisturbed soils and corrosion in disturbed soils. Corrosion in undisturbed soil is always low, regardless of soil conditions, and is limited only by the availability of the oxygen necessary for the cathodic reaction. Steel piles driven into soil fall under this category and therefore undergo limited corrosive attack. Corrosion of metals in disturbed soils is strongly affected by soil conditions. Soil electrical resistivity, mineral composition, dissolved salts, moisture content, total acidity or alkalinity (pH), redox potentials, microbiological activity, and concentration of oxygen play important roles in determining whether metal corrosion is a serious problem in soils ( Table ). Other, secondary factors are also important but are more difficult to define. Thus, simply testing metals and alloys in a variable pH solution or in aerated or deaerated solutions will not accurately simulate the conditions in soil (Ref). Any metal buried by backfilling is in a disturbed soil and is subject to corrosion attack, depending on the characteristics of the soil. Most metals in soil are exposed to disturbed-soil conditions. Because it is the more common of the two, only the disturbed-soil condition is discussed.