In addition to soil nutrient deficiencies, poor production from bermudagrass [Cynodon dactylon (L.) Pers.] hay fields may result from soil compaction. The objective of this study was to quantify ‘Coastal’ bermudagrass dry matter (DM) yield responses to N source, K application, and subsoil tillage. The study was conducted during 3 yr at the Brown Loam Experiment Station, Raymond, MS, on a Loring silt loam soil (fine-silty, mixed, thermic Typic Fragiudalfs). A randomized complete block design with four replicates of a 3 × 2 × 2 factorial arrangement was used. Three levels of N source, zero N, and ammonium nitrate (AN) or ammonium sulfate (AS) at 120 kg N ha^-1 yr^-1, and two levels of K (0 and 60 kg K₂O ha^-1 yr^-1 as KCl) were applied to subsoil-tilled vs. untilled plots.  Annual yield response to fertilizer treatments followed similar patterns each year.  Application of K did not have any effect on yield.  Both sources of N produced similar yields (AN, 7080 kg ha^-1 in 2001, 7900 kg ha^-1 in 2002, and 8760 kg ha^-1 in 2003; AS, 7160 kg ha^-1 in 2001, 7650 kg ha^-1 in 2002, and 8660 kg ha^-1 in 2003), which were greater than when no N was applied (5630 kg ha^-1 in 2001, 4390 kg ha^-1 in 2002, and 4830 kg ha^-1 in 2003).  In 2001, subsoil-tilled plots had lower yield (6260 kg ha^-1) than untilled plots (6980 kg ha^-1).  No differences were observed between tillage treatments in subsequent years.  These results suggest that lack of N fertilizer can be detrimental to bermudagrass hay production, but there was no yield advantage from application of K.  Any benefits from subsoil tillage seemed to be inconsistent on relatively flat silt loam soils.