Soil tests of topsoil (0"-6") and subsoil (6"-12" and 12"-18") samples have previously identified a subsoil potassium deficiency on considerable acreage over the Mississippi Delta cotton soils. This paper presents the first two years of a four year study of the effect of potassium rates and placement systems on lint yield of four varieties grown on a potassium deficient soil with 5.9 pH values or higher.

In an effort to develop a more economical and efficient means of correcting subsoil potassium deficiencies, deep band placement in the drill behind a parabolic super chisel shank was explored. Three fertilizer rates (80, 120, and 160 lb K₂ O/A) and three placement systems (surface, deep banded and split 1/2 surface and 1/2 deep banded) were applied to four varieties (DES 119, DP 50, St 453, and C-130) to study their effect on lint yield.

All nine potassium fertilizer treatments increased lint yields over the check treatment when averaged over varieties. The responses to deep banding and to split application of potassium were most effective with DES 119 and least effective with St 453. DES 119 produced the highest (P<.05) lint yields of the four varieties. Only one treatment, split application of 160 lb K₂ O/A, produced higher (P<.05) lint yields over all varieties than all three surface (80, 120, and 160 lb K,O/A) treatments. The mean yields for the 160 lb K₂ O/A rate was higher (P<.05) than for either the 80 or 120 lb K₂ O/A rates. Overall, the mean lint yields for both split and deep application systems were significantly higher (P<.05) than the surface broadcast incorporated system.