AbstractAcquisition of mineral nitrogen by roots from the surrounding environment is often not completely efficient, in which a variable amount of leakage (efflux) relative to gross uptake (influx) occurs. The efflux/influx ratio (E/I) is, therefore, inversely related to the efficiency of nutrient uptake at the root level. Time‐integrated estimates of E/I and other nitrogen‐use traits may be obtainable from variation in stable isotope ratios or through compartmental analysis of tracer efflux (CATE) using radioactive or stable isotopes. To compare these two methods, Populus balsamifera L. genotypes were selected, a priori, for high or low nitrogen isotope discrimination. Vegetative cuttings were grown hydroponically, and E/I was calculated using an isotope mass balance model (IMB) and compared to E/I calculated using 15N CATE. Both methods indicated that plants grown with ammonium had greater E/I than nitrate‐grown plants. Genotypes with high or low E/I using CATE also had similarly high or low estimates of E/I using IMB, respectively. Genotype‐specific means were linearly correlated (r = 0.77; P = 0.0065). Discrepancies in E/I between methods may reflect uncertainties in discrimination factors for the assimilatory enzymes, or temporal differences in uptake patterns. By utilizing genotypes with known variation in nitrogen isotope discrimination, a relationship between nitrogen isotope discrimination and bidirectional nitrogen fluxes at the root level was observed.