Passive air samplers (PASs) are simple, versatile devices that are increasingly used to determine the concentrations of semivolatile organic compounds (SVOCs) in the atmosphere. Using PAS and interpreting PAS-derived data with confidence requires a detailed understanding of the factors that control the uptake kinetics. A number of experiments were aimed at clarifying the role that the housing has in limiting the uptake of SVOCs in a PAS. Specifically, we quantified the gradient in the amount of polychlorinated biphenyls (PCBs) accumulated in XAD-filled mesh cylinders with increasing distance from the PAS housing's opening. That gradient was non-existent in an artificially ventilated housing (i.e. different segments of a cylinder contained the same amount of PCBs), minor during outdoor deployments (i.e. the bottom third of the cylinder sampled approximately 20% more PCBs than the top third), and strong during indoor deployments (i.e. the bottom third of the cylinder sampled twice the amount sampled by the top third). This is consistent with the thickness of the air boundary layer surrounding the XAD-resin increasing with increasing distance from the housing's opening and decreasing with increasing air turbulence. An experiment with housings absorbing different amounts of sunlight revealed that heat-induced convection has a minor effect on the gradient within the mesh cylinder and on the total amount of accumulated PCB. Similarly, this gradient and the total amount sorbed was also not influenced by the number of XAD-filled mesh cylinders placed within a housing as long as they were deployed outdoors. However, if four mesh cylinders were placed in one housing in a calm indoor setting, the top third of the mesh cylinders was notably starved of PCBs, suggestive of an air concentration gradient within the sampler housing.