Woody debris (WD) is an important component of forest C budgets, both as a C reservoir and source of CO2 to the atmosphere. We used an infrared gas analyzer and closed dynamic chamber to measure CO(2) efflux from downed coarse WD (CWD; diameter>or=7.5 cm) and fine WD (FWD; 7.5 cm>diameter>or=2 cm) to assess respiration in a selectively logged forest and a maturing forest (control site) in the northeastern USA. We developed two linear regression models to predict WD respiration: one based on WD temperature, moisture, and size (R2=0.57), and the other on decay class and air temperature (R2=0.32). WD respiration (0.28+/-0.09 Mg C ha-1 year-1) contributed only approximately 2% of total ecosystem respiration (12.3+/-0.7 Mg C ha-1 year-1, 1999-2003), but net C flux from CWD accounted for up to 30% of net ecosystem exchange in the maturing forest. C flux from CWD on the logged site increased modestly, from 0.61+/-0.29 Mg C ha-1 year-1 prior to logging to 0.77+/-0.23 Mg C ha-1 year-1 after logging, reflecting increased CWD stocks. FWD biomass and associated respiration flux were approximately 7 times and approximately 5 times greater, respectively, in the logged site than the control site. The net C flux associated with CWD, including inputs and respiratory outputs, was 0.35+/-0.19 Mg C ha-1 year-1 (net C sink) in the control site and -0.30+/-0.30 Mg C ha-1 year-1 (net C source) in the logged site. We infer that accumulation of WD may represent a small net C sink in maturing northern hardwood forests. Disturbance, such as selective logging, can enlarge the WD pool, increasing the net C flux from the WD pool to the atmosphere and potentially causing it to become a net C source.