AbstractSoil salinity quickly became a problem in sheltered lysimeters used for studies of the drainage requirements of crops in North Carolina. Water tables were maintained by subsurface watering at depths of 6 to 40 inches below the soil surface in 3‐foot‐square lysimeters, with some sheltered from rainfall and others exposed to rainfall. All were exposed alike to other atmospheric conditions. Very high levels of P, K, and Ca were established in the soil by mixing 0‐20‐0, muriate of potash, and lime with the soil placed into each tank.Soybean plants exhibited chlorosis and stunted growth in the lysimeters with subsurface watering only. These symptoms were serious approximately 1 month after the establishment of water tables only 6 inches beneath the surface, but were less pronounced with the deeper water tables. They were not observed where rain fell on the soil.At any water table depth, the plant growth and yield were best on the lysimeters that were leached with rainfall. Plant growth increased with depth to water table to a depth of 24 inches, both with and without rain on the soil.After soybean harvest in 1959, the electrical conductivity of the saturation extract from the soil in the top 6 inches of all lysimeters was 4 to 5 millimhos where subsurface watering only was used, and 0.4 millimho where rain fell on the soil.Lower fertilizer applications and leaching with small amounts of water kept the soil salinity in the upper 6‐inches at a low level in subsequent years. Crops of soybeans, cabbage, corn, grain sorghum, and string beans were produced successfully both in the lysimeters sheltered from rain and in those exposed to rain.