A great deal of interest and emphasis has been placed on cotton (Gossypium spp.) fiber quality, as well as yield benefits associated with potassium (K) fertilization. Due to limited information describing the response of cotton in Arizona to K fertilization, three studies were conducted each in 1992 and 1993 with the objective of evaluating the response of cotton crop development and lint yield to soil and/or foliar applications of K fertilizer. The 1992 locations of the trials included the Safford Agricultural Center (Pima clay loam), Maricopa Agricultural Center (Casa Grande sandy loam) and a site at Coolidge (Mohall sandy loam). The 1993 sites included Safford, Maricopa Agricultural Centers and in the Yuma Valley Gadsden clay. All irrigation, pest management, and fertilization inputs (other than K), were administered in accordance with the optimum required levels throughout the season. Routine plant measurements and mapping analyses were carried out at each location on regular 14 day intervals throughout the season. At the Safford location, both Upland (G. hirsutum L., var DPL 90) and Pima (G. barbadense. L., var S-6) cotton were planted with treatments including both soil and foliar K applications. Soil applications were broadcast and preplant incorporated using K₂SO₃ as the K source at rates of 0, 200, and 400 lbs K₂O/acre. Four 4.6 lbs K₂O/acre foliar applications of KNO₃ were applied at two-week intervals over the fruiting period. The trial at the Maricopa Agricultural Center (Pima S-7) and the Yuma Valley (DPL 5409) included four foliar K applications over the growing season. The foliar treatments included rates which ranged from 0 to 37 lbs K₂O/acre using KNO₃ as the K source. At Coolidge all K treatments were band applied to the soil at a depth of 8 in., preplant on Upland (STV KC 311). The treatments were 0, 218, 436, and 654 lbs K₂O/acre using K₂SO₃ as the K source. Treatments were arranged over the experimental area in a randomized complete block design with five replications. Results from trials for each year indicated no differences among any of the treatments (including soil and/or foliar and unfertilized treatments). All of the plant measurements taken from all the locations for the two years revealed good crop growth resulting in excellent fruit retention without excessive vegetative growth (ie. height-to-node ratios within the 95% confidence intervals for both Upland and Pima cotton). This indicated ample nutrient demand so that if available soil K was inadequate to meet crop needs, deficiency symptoms and reduced yields should have occurred. There were no visual symptoms detected from any treatments in the experiments for the two years (all locations). The results of these K fertility experiments support current University of Arizona recommendations that unless exchangeable K is less than 150 ppm, crop response to K fertilization is not likely.