With the objectives of analyzing N recovery and potential N losses in the warm-season hybrid bermudagrass 'Tifgreen' [Cynodon dactylon (L.) Pers. x C. transvaalensis Burtt-Davy], two greenhouse studies were conducted. Plugs were planted in PVC cylinders filled with a modified sandy growing medium. Urea (URE), sulfur-coated urea (SCU), and Hydroform (HYD) (Hydro Agri San Francisco, Redwood City, CA) were broadcast at rates of 100 and 200 kg N ha-1 every 20 and 40 d. The grass was clipped three times every 10 d and analyzed for N concentration and N yield. In addition, leachates were analyzed for NO3-N. Use of the least soluble source, HYD, resulted in the lowest average clipping N concentration and N yield, as compared with SCU and URE. Clipping N concentration and N yield showed a cyclic pattern through time, particularly under long-day (> 12 h) conditions. When the photoperiod decreased below 12 h, leachate NO3-N concentration exceeded the standard limit for drinking water (10 mg L-1) by 10 to 19 times with the high SCU and URE application rate and frequency. However, leaching N losses represented a minimal fraction (< 1%) of the total applied N. More applied N was recovered in plant tissues using SCU and URE (89.5%) than using HYD (64.1%), with more than 52% of applied N accumulating in clipping. Highly insoluble N sources such as HYD decrease N leaching losses but may limit bermudagrass growth and quality. Risks of NO3-N losses in bermudagrass can be avoided by proper fertilization and irrigation programs, even when a highly soluble N source is used.