A series of field studies was conducted on Alabama soils to evaluate cotton response to surface broadcast and deep placement of K fertilizer. Experiments were initiated in 1989 on a Emory silt loam and a Norfolk sandy loam, and in 1990 on a Lucedale sandy clay loam. Potassium was applied at rates ranging from 0 to 90 lb K₂O A^-1. The K was either broadcast on the surface or deep placed. Surface broadcast applications were made with and without in-row subsoiling. Deep placement was achieved with a dry fertilizer applicator which applied the dry fertilizer at depths of 6 to 15 inches behind a subsoiler shank. The Emory and Lucedale soils also received 120 lb K,O A-1 deep placed. Two additional treatments for all three soils received 1500 lb agricultural limestone with and without 90 lb K₂O A^-1. During the test the only site where there was a significant difference for surface versus deep placement treatments was on the Norfolk soil. On the Norfolk soil, deep placement of 30 lb K₂O A^-1 produced higher yields as compared to the surface Broadcast application of 30 lb K₂O A^-1. At higher rates the surface broadcast treatments consistently produced higher yields as compared to the deep placement treatments. For the Emory and Lucedale soils, there were no significant differences between the two methods of application. Results of these field studies suggest that for Alabama soils, the deep placement of K fertilizer for cotton is not superior to surface broadcast applications of K. The results also show that the deep placement of agricultural limestone with and without K fertilizer for cotton is not justified.