The effects of nitrogen (N) and potassium (K) fertilization are being evaluated in a long-term continuous cotton production area without the aid of irrigation.  A 3 x 2 factorial arrangements of N rates (101, 134, and 168 kg N ha^-1 and K rates (0 and 67 kg K ha^-1) have been included in a randomized complete block design with ten replications.  The fertilizer materials are applied as urea-ammonium nitrate solution (32% N) and muriate of potash solution (0-0-16, 155.8 g K l^-1).  The study was initiated in 2001 on a Forestdale/Dundee silty clay loam (Typic Endoaqualfs) at the Tribbett Satellite Farm of the Delta Research and Extension Center.  In the first years of the study there was no response to either N rate or K rate with yields related more to water availability than to nutrient levels.  With continued cotton production, 2003 yields were significantly increased with increasing N rates with the top yield, 4066 kg seedcotton ha^-1 produced with 168 kg N ha^-1 (1634 kg lint ha^-1).  The 2004 yields were similar to those achieved in 2003 with the top yield (1747 kg lint ha^-1) produced with 168 kg N ha^-1.  The 2005 yields were lower than those measured in previous years largely due to weather effects from the hurricanes that sent high winds and rainfall through the area and knocked out cotton.  Yields were affected by both N rate and K rate indicating some response to K buildup in the field.  The 2006 yields were greatly reduced by the effects of drought that were prevalent during the growing season.  Seedcotton yields were generally less than 1750 kg ha^-1 for the season.  While there was no response to increasing N rates, there was a significant increase in yield with respect to added K fertilizer.