Potassium fertilization is critical to produce high-yielding rice (Oryza sativa L.) and soybean [Glycine max (Merr.) L.] grown on poorly buffered silt loams in Arkansas. The research objectives were to evaluate the influence of annual K fertilization rate on rice and soybean yields and soil-test K across time. Muriate of potash was applied at 0, 30, 60, 90, and 120 kg K ha^-1 annually to a Calhoun silt loam cropped with a rice-soybean rotation beginning in 2000. The same K rate was applied to each plot annually with all other crop inputs maintained at an optimum level. Plant tissue samples were collected annually at the R2 stage of soybean or the panicle differentiation and late boot stages of rice. Soil samples were collected from each plot before the initial K application and annually thereafter from the 0-to 10-cm depth and extracted with Mehlich-3 to monitor soil-test K across time. The recommended K-fertilizer rates for the unfertilized control ranged from 60 to 120 kg K ha^-1 depending on the year and crop. Soil-test K in the unfertilized control fluctuated among years ranging from 68 mg K kg^-1 in 2005 to 107 mg K kg^-1 in 2003. Depending on the year, soil-test K increased linearly by 1 mg kg^-1 for each 4 to 11 kg K ha^-1. During the first two rice-soybean rotation cycles, only soybean yields in 2001 were significantly increased by K fertilization. During the third rotation cycle, rice and soybean yields were both significantly increased by K fertilization with the greatest yields produced by 120 kg K ha^-1. Yield increases ranged from 7-15% for rice in 2004 and 15-37% for soybean in 2005. Annual application of recommended K rates is needed to maintain adequate soil K fertility to achieve crop yield potential.