During the Program for Research on Oxidants: PHotochemistry, Emissions, and Transport (PROPHET) campaign from 21 July to 3 August 2016, field experiments on leaf-level trace gas exchange of nitric oxide (NO), nitrogen dioxide (NO2), and ozone (O3) were conducted for the first time on the native American tree species Pinus strobus (eastern white pine), Acer rubrum (red maple), Populus grandidentata (bigtooth aspen), and Quercus rubra (red oak) in a temperate hardwood forest in Michigan, USA. We measured the leaf-level trace gas exchange rates and investigated the existence of an NO2 compensation point, hypothesized based on a comparison of a previously observed average diurnal cycle of NOx (NO2+NO) concentrations with that simulated using a multi-layer canopy exchange model. Known amounts of trace gases were introduced into a tree branch enclosure and a paired blank reference enclosure. The trace gas concentrations before and after the enclosures were measured, as well as the enclosed leaf area (single-sided) and gas flow rate to obtain the trace gas fluxes with respect to leaf surface. There was no detectable NO uptake for all tree types. The foliar NO2 and O3 uptake largely followed a diurnal cycle, correlating with that of the leaf stomatal conductance. NO2 and O3 fluxes were driven by their concentration gradient from ambient to leaf internal space. The NO2 loss rate at the leaf surface, equivalently the foliar NO2 deposition velocity toward the leaf surface, ranged from 0 to 3.6 mm s−1 for bigtooth aspen and from 0 to 0.76 mm s−1 for red oak, both of which are ∼90 % of the expected values based on the stomatal conductance of water. The deposition velocities for red maple and white pine ranged from 0.3 to 1.6 and from 0.01 to 1.1 mm s−1, respectively, and were lower than predicted from the stomatal conductance, implying a mesophyll resistance to the uptake. Additionally, for white pine, the extrapolated velocity at zero stomatal conductance was 0.4±0.08 mm s−1, indicating a non-stomatal uptake pathway. The NO2 compensation point was ≤60 ppt for all four tree species and indistinguishable from zero at the 95 % confidence level. This agrees with recent reports for several European and California tree species but contradicts some earlier experimental results where the compensation points were found to be on the order of 1 ppb or higher. Given that the sampled tree types represent 80 %–90 % of the total leaf area at this site, these results negate the previously hypothesized important role of a leaf-scale NO2 compensation point. Consequently, to reconcile these findings, further detailed comparisons between the observed and simulated in- and above-canopy NOx concentrations and the leaf- and canopy-scale NOx fluxes, using the multi-layer canopy exchange model with consideration of the leaf-scale NOx deposition velocities as well as stomatal conductances reported here, are recommended.