Significance and Use
5.1 The crumb test method provides a simple, quick method for field or laboratory identification of a dispersive clay soil. The internal erosion failures of a number of homogeneous earth dams, erosion along channel or canal banks, and rainfall erosion of earthen structures have been attributed to colloidal erosion along cracks or other flow channels formed in masses of dispersive clay (5).
5.2 The crumb test method, as originally developed by Emerson (6), was called the aggregate coherence test and had seven different categories of soil-water reactions. Sherard (5) later simplified the test by combining some soil-water reactions so that only four categories, or grades, of soil dispersion are observed during the test. The crumb test is a relatively accurate positive indicator of the presence of dispersive properties in a soil. The crumb test, however, is not a completely reliable negative indicator that soils are not dispersive. The crumb test can seldom be relied upon as a sole test method for determining the presence of dispersive clays. The double-hydrometer test (Test Method D4221) and pinhole test (Test Method D4647) are test methods that provide valuable additional insight into the probable dispersive behavior of clay soils.
1.3 The crumb test method has some limitations in its usefulness as an indicator of dispersive clay. A dispersive soil may sometimes give a non-dispersive reaction in the crumb test. Soils containing kaolinite with known field dispersion problems have shown non-dispersive reactions in the crumb test (1)2. However, if the crumb test indicates dispersion, the soil is probably dispersive.
1.4 Oven-dried soil should not be used to prepare crumb test specimens, as irreversible changes could occur to the soil pore-water physicochemical properties responsible for dispersion (2).
1.5 The crumb test method, while a good quick indication of dispersive clay, should usually be run in conjunction with a pinhole test and a double hydrometer test, Test Methods D4647 and D4221, respectively.
1.6 All observed and calculated values shall conform to the guidelines for significant digits and rounding established in Practice D6026.