The irregular shape, small size, and spotty occurrence of sodium‐affected soils (NaAS) in moderately thick to thin loess over Illinoian till have continued to slow mapping in south‐central Illinois. Use of a Geonics EM‐38 ground conductivity meter to recognize soils with natric horizons in Tennessee suggested that this device would be useful in Illinois. We selected eight areas of NaAS in five counties of south‐central Illinois to find if the EM‐38 would be useful for predicting exchangeable sodium percentage (ESP). Approximately 225 observation points, spaced at 10‐m intervals, were established in each area and 107 sample cores were collected for laboratory analyses. Weighted averages of the ESP, electrical conductivity of the saturation extract from a saturated soil paste (ECe), and field pH data for 0 to 75‐ and 75 to 150‐cm depths were compared with the horizontal (EMh) and vertical (EMv) responses and the mean ESP data increases with depth, i.e., 7.0 at 35 cm, 9.7 at 75 cm, and 12.5 at 135 cm. The EMh readings correlate almost as well with ESP (r = 0.73) as with ECe (r = 0.76) for the 0‐ to 75‐cm depths. The EMv readings correlate slightly better with ESP (r = 0.76) than with ECe (r = 0.68) for the 75‐ to 150‐cm depth. These results show that EM readings may by used to predict ESP. The correlations of the depth‐weighted average ESP at the maximum depth for meeting natric horizon criteria in these soils (75 cm), and the EMv readings show that the EM‐38 provides a rapid and accurate method to describe the composition of the map units of NaAS on moderately thick to thin loess over Illinoian till in south‐central Illinois.

A 4‐yr study was conducted to determine corn (Zea mays L.) and soybean [Glycine max (L.) Merr.] yield and plant population differences between three south central Illinois soils with varying levels of exchangeable Na in subsoils. The 2‐yr corn yield average was highest (190 bu/acre) for Cisne soils, which had low exchangeable Na in the subsoil, and lowest (118 bu/acre) for the Huey soils, which had the highest level of Na in the subsoil. The 2‐yr corn plant populations at harvest were not significantly reduced by exchangeable Na in the subsoil. The 2‐yr average soybean yields were less affected by Na; however, the 1991 soybean yield was significantly reduced by 20 bu/acre. The Cisne soils averaged 43 bu/acre while the Huey soils, with higher exchangeable Na levels, had lower yields (35 bu/acre). The 2‐yr soybean plant populations at harvest were not significantly reduced by exchangeable Na in the subsoil. The corn yield (160 bu/acre) was reduced for Cisne tax‐adjunct soils, which had a medium level of exchangeable Na but there was no corresponding reduction in the soybean yield (44 bu/acre). Productivity index ratings for soils with varying levels of exchangeable Na should reflect this adverse crop yield response.