Sporadic potassium deficiency symptoms have recently been reported throughout the cotton belt. Some researchers have speculated that this phenomenon is connected to the development and use of early maturing, fast fruiting, and high yielding genotypes. The hypothesis was that the root systems of the fast fruiting genotypes were not able to satisfy the high K demand of the developing bolls. In this study, our objective was to investigate genotypic variability in lint yield and fiber quality as affected by potassium and nitrogen fertilization. We also sought to determine if there was an interaction between varying levels of K and N fertilization on lint yield and fiber quality.

A field study was conducted in 1991 on eight cotton genotypes representing a range of maturity and regional adaptations, DES 119, DPL 5415, HS 26, MD 51 ne, PeeDee 3, STV 453, STV 825, and STV 887. All plots received a preplant application of 110 kg ha^-1 N. Half of these plots also received a 40 kg ha^-1 sidedress application of N. Within each N treatment, half the plots received a preplant incorporated surface application of 135 kg ha^-1 K, with the remaining half receiving no K. Experimental design was a split-split plot with N levels as the main plot, K levels as the subplot and genotypes as the sub-subplot. Two above ground dry matter harvests were collected on DES 119 and MD 51 ne, one just prior to the sidedress N application and the other occurring at approximately cutout.

Plant growth was altered by K deficiency. The low K plants had an 18 % lower leaf area index (LAI) and 12 % lower stem weight. K deficiency increased the specific leaf weight (SLW) by 16 %, however. Supplemental N caused DES 119 to allocate more dry matter to vegetative growth. MD 51 ne did not alter dry matter partitioning in response to supplemental N. Lint yield was reduced 117 kg ha^-1 and lint % decreased 2 % in the K deficient plants as compared to K fertilized plants. K deficiency produced a 7 % decrease in fiber MIC and a 4 % reduction in fiber maturity. Fiber perimeter was not influenced by K fertilization. In general, no K-by-genotype or N-by-genotype interaction was detected for any of the yield or fiber quality components. In addition, N did not interact with K to affect any parameters. The data from this study imply that the genotype a producer should use under K deficient conditions is the same genotype the producer would choose for K sufficient conditions.