Nitrogen (N) release from legume cover crops is a key N source for subsequent crops in rotation. In this study, chopped fresh shoots or roots (<5 mm) of crimson clover (CC), hairy vetch (HV), and red clover (RC) were incorporated into a 50:50 mixture of air-dried sandy loam soil (<2 mm) and washed builders sand at a rate of 300 mg N kg^-1. The mixtures were packed in leaching tubes (four replicates), leached with 100 mL of 5 mmol L^-1 CaCl₂, and then incubated for 10 wk (22 °C, 0.33 bar matric potential) with weekly leaching. Total N and inorganic N (NH₄⁺ plus NO₃^-) in leachate were quantified and organic N was determined as the difference between total N and inorganic N. More N was released from shoots (63.4%–70.0% of initial N) than from roots (27.3%–50.7% of initial N). Mineralized organic N and inorganic N followed the first order, single N-pool mineralization model [N_t = N₀(1 – e^-kt); R² = 0.94-0.99]. Potentially mineralizable N (N₀, as % of initial N) was similar for shoots (CC = 75.1%, HV = 74.2%, and RC = 71.3%), but varied for roots (CC = 36.2%, HV = 52.6%, and RC = 53.0%). The N₀ pool in shoots had a half-life (t_1/2 = ln 2/k) of 11.0, 9.8, and 15.1 d for CC, HV, and RC, respectively; and a half-life in roots of 23.9, 8.5, and 25.7 d, respectively. Hence, HV released its stored N in both roots and shoots faster than CC and RC. The results in this study would help farmers optimize their choice in legume cover crops and termination times to better synchronize N release with crop uptake.