Root length density of maize ( Zea mays L.) and soybean Glycine max (L.) Merr., measured with minirhizotrons, was compared with soil-water depletion, measured with neutron probes, to determine whether the soil-water-depletion method could indicate root length density and root depth in a stratified soil. The study was carried out in southeast Kansas, USA, on a Parsons silt loam soil (Mollic Albaqualf fine, mixed, thermic) overlying clay (a claypan). Measurements were taken during two dry periods, one in 1983 and one in 1984, to a depth of 1.35 m for soil water and to a depth of 1.15 m for root length density. At the beginning of the dry period, average soil water content for both crops throughout the profile was 0.47 m 3 m −3 in 1983 and 0.42 m 3 m −3 in 1984. During both dry periods, roots died back in the surface soil because the silt loam dried out; it was drier than the clay. The claypan inhibited root growth. In the clay, roots of both crops increased in root length density in the water year (1983) or decreased in root length density in the drier year (1984). In both years, the depth of maximum water depletion in the soil profile for maize (0.45 m in 1983, 0.90 m in 1984) corresponded well with the depth of maximum root length density measured with minirhizotrons (0.50 m in 1983, 0.80 m in 1984). The depths of maximum depletion for soybean (0.15 m and 0.30 m in 1983, and 0.60 m in 1984) also corresponded relatively well with the zones of maximum root length density (0.15 and 0.50 m in 1983, and 0.40 m in 1984). Depth of rooting was not evident from the soil-water measurements. Root length densities for both crops were low (less than about 1000 and 300 mm −3 for maize and soybean, respectively) at bottom depths of the profile.