Cotton response to surface applications of potassium fertilizer: A 10-yearsummary

Field studies were conducted for 10 yr on a Limestone Valley soil (Dewey si lt loam, a fine, kaolinitic, thermic Typic Paleudults) in north Alabama to evaluate cotton (Gossypium hirsutum L.) yield and lint quality response to long-term surface applications of K. Potassium rates were 0, 60, 120, and 1 80 lb K2O/acre for the first 3 yr of the study. Fertilizer was broadcast in the spring or split applied in fall and spring. No K fertilizer was applie d during the fourth year (1990). After the fourth year, half of the treatme nts were left in residual and remaining treatments continued to receive ann ual applications of K (1991-1996). The experiment was a split plot arrangem ent of a randomized complete block. From 1987 to 1989, two varieties were c ompared, which served as whole plots. From 1991 through 1996, one variety w as grown and annual vs. residual K fertilization served as whole plots. Pot assium treatments were used as subplots throughout the test. From 1987 to 1 989 there were no differences between 'Stoneville 825' and 'Deltapine 50'. Lint yields were increased significantly by K fertilization in 6 of 9 yr th at cotton was grown on this low K soil. Potassium fertilization increased l int yields by as much as 450 lb/acre in responsive years. From 1991 through 1996, residual K produced lesser yields than where K was applied annually. There were no differences in lint yield among split (fall-spring), fail, o r spring applied K. in responsive years, micronaire increased with K rate. Our results support the calibration currently used by the Auburn University Soil Testing Lab for Limestone Valley soils.