Despite its importance in the terrestrial C cycle rhizosphere carbon flux (RCF) has rarely been measured for intact root–soil systems. We measured RCF for 8-year-old saplings of sugar maple (Acer saccharum) and yellow birch (Betula allegheniensis) collected from the Hubbard Brook Experimental Forest (HBEF), NH and transplanted into pots with native soil horizons intact. Five saplings of each species were pulse labeled with 13CO2 at ambient CO2 concentrations for 4–6 h, and the 13C label was chased through rhizosphere and bulk soil pools in organic and mineral horizons for 7 days. We hypothesized yellow birch roots would supply more labile C to the rhizosphere than sugar maple roots based on the presumed greater C requirements of ectomycorrhizal roots. We observed appearance of the label in rhizosphere soil of both species within the first 24 h, and a striking difference between species in the timing of 13C release to soil. In sugar maple, peak concentration of the label appeared 1 day after labeling and declined over time whereas in birch the label increased in concentration over the 7-day chase period. The sum of root and rhizomicrobial respiration in the pots was 19% and 26% of total soil respiration in sugar maple and yellow birch, respectively. Our estimate of the total amount of RCF released by roots was 6.9–7.1% of assimilated C in sugar maple and 11.2–13.0% of assimilated C in yellow birch. These fluxes extrapolate to 55–57 and 90–104 g C m−2 yr−1 from sugar maple and yellow birch roots, respectively. These results suggest RCF from both arbuscular mycorrhizal and ectomycorrhizal roots represents a substantial flux of C to soil in northern hardwood forests with important implications for soil microbial activity, nutrient availability and C storage.