In chronic toxicity tests with Caenorhabditis elegans, it is necessary to feed the nematode with bacteria, which reduces the freely dissolved concentration (Cfree) of hydrophobic organic chemicals (HOCs), leading to poorly defined exposure with conventional dosing procedures. We examined the efficacy of passive dosing of polycyclic aromatic hydrocarbons (PAHs) using silicone O-rings to control exposure during C. elegans toxicity testing and compared the results to those obtained with solvent spiking. Solid-phase microextraction and liquid-liquid extraction were used to measure Cfree and the chemicals taken up via ingestion. During toxicity testing, Cfree decreased by up to 89% after solvent spiking but remained constant with passive dosing. This led to a higher apparent toxicity on C. elegans exposed by passive dosing than by solvent spiking. With increasing bacterial cell densities, Cfree of solvent-spiked PAHs decreased while being maintained constant with passive dosing. This resulted in lower apparent toxicity under solvent spiking but an increased apparent toxicity with passive dosing, probably as a result of the higher chemical uptake rate via food (CUfood). Our results demonstrate the utility of passive dosing to control Cfree in routine chronic toxicity testing of HOCs. Moreover, both chemical uptake from water or via food ingestion can be controlled, thus enabling the discrimination of different uptake routes in chronic toxicity studies.