Although the fate of fertilizer applied to turfgrass has been studied in the past, recovery of applied fertilizer N is typically low, and denitrification has been cited as the reason. The objectives of this research were twofold: (i) to examine the fate of 15N applied to Kentucky bluegrass (Poa pratensis L.) turf as KNO3, including direct measurement of denitrification; and (ii) to determine whether and how plants affect fertilizer-N recovery. Polyvinyl chloride (PVC) cylinders, modified to permit atmospheric sampling, were used throughout field experiments during the spring and summer 1999 and a greenhouse experiment in 2000. Potassium nitrate (98.5 atom % 15N) was applied in solution at 49 kg N ha-1 to replicated plots, and atmospheric samples were collected three times a day from 0800 to 1100, 1100 to 1400, and 1400 to 1700 h during a 6-wk period in the spring and a 4-wk period during the summer of 1999. Emission of N2 or N2O ranged from 3.3 to 21.3% and from 0.3 to 5.9% of labeled fertilizer N (LFN), respectively. Recovery of LFN in the soil or plant, plus that emitted as N2 or N2O, ranged from 57.4 to 73.2%. A 4-wk greenhouse experiment comparing LFN recovery for bare soil and turf, including gas emission and leachate, was initiated in the summer of 2000. Total emission of LFN as N2 or N2O was 19.0% for the turfgrass, as compared with 7.3% for the bare soil. Corresponding values for total recovery of LFN were 70.6 and 84.2%, respectively.