Polyaromatic hydrocarbons (PAHs) are complex carbonaceous compounds emitted to the atmosphere by various combustion processes. Because the toxicity of many of them is now well recognized and documented, the determination of their atmospheric concentrations is of great interest to better understand and develop future atmospheric pollution control strategies. Hence, a common sampling protocol has to be defined to homogenize the results. With this goal in mind, field studies were carried out under different environmental conditions (74 samples) by simultaneously operating both a conventional sampler and a sampler equipped with a denuder tube upstream from the filter. The experimental results presented in this work show that the atmospheric particulate PAH concentrations are underestimated at least by a factor of 2 using a conventional sampler. The discrepancy between the two kinds of samplers used varied a lot from one compound to another and from one field campaign to another. This discrepancy may be explained by a simple degradation of particulate PAH in the natural atmosphere and on the filter. This is particularly worrisome because, based on the results presented in this work, the atmospheric PAH concentrations measured using conventional samplers not equipped with an ozone trap can underestimate the PAH concentration by more than 200%. This is especially true when the samples are collected in the vicinity of the point source of particulate PAHs and for highly reactive compounds such as benzo[a]pyrene.