Routine sampling and analysis of soil from agronomic production fields is the standard process of determining the nutrients and rates needed to optimize crop yield and/or profit.  The acceptance and implementation of grid soil sampling practices has affected how and when soil samples are collected, as well as soil-test laboratory management.  Information from the University of Arkansas Marianna Soil Test Laboratory was used to examine trends in the number of soil samples analyzed, sample type (grid or field average), and Mehlich-3 extractable P and K from samples that were analyzed from January 2006 through March 2011.  Since 1976, the total number of soil samples analyzed has increased linearly (r² = 0.88, slope = 3062 samples/year).  Over the last 10 years, however, the number of samples analyzed has increased by 6990 samples/year (r² = 0.78) due largely to grid soil sampling.  In 2006, grid soil samples accounted for only 18% of client samples analyzed, but by 2010, had increased to 61% of all samples.  The majority (80-85%) of soil samples are analyzed from October through April.  Although the percentage of samples analyzed by month has been static, the actual number of samples analyzed has increased.  Median annual soil test P and K values from 2006-2010 for specific cropping systems shows that soil P availability is static for most row crop soils, but is decreasing rapidly (7 to 14 ppm/year) in soils used to produce cool- and warm-season grasses (hay).  Median soil test K was decreasing for all crops examined but was greatest in soils used for corn and cotton (8.5 ppm/year) and hay production (12 to 16 ppm/year).  Median soil test P and K were found to decline linearly from October through March for soil samples collected following cotton and curvilinearly for soil samples collected following soybean.