Recent studies have shown that a reverse flow often occurs in a unidirectional airflow in push-pull ventilation and may transport contaminants from the source into a worker's breathing zone. The same problem may arise in local exhaust ventilation when the contaminant source is located in the worker's wake region. In this study, organic solvent work with local exhaust ventilation was duplicated in a laboratory and the details of the reverse flow around the worker's body produced by the ventilation were experimentally investigated. In order to evaluate the influence of the reverse flow on the exposure of the worker, experiments with a mock-up mannequin (dummy worker) and a local ventilation system which was equipped with an exterior type hood and an enclosure type hood were conducted. The exposure level and the contaminant leakage from the hoods in several conditions were measured by means of a smoke test and tracer gas method. Ethanol vapor was used as a tracer gas. With the exterior type hood, the reverse flow visualized by the smoke was observed in front of the standing dummy worker but could not be observed when the dummy worker was seated. From the tracer gas measurements, it was proved that the exposure due to the reverse flow was not so serious at a capture velocity of > 0.4 m/s, but < 10 ppm contaminant leakage from the exterior hood had been recognized independently of the capture velocity. With the enclosure type hood, exposure due to the reverse flow could be controlled with a capture velocity of > 0.8 m/s. Although the contaminant leakage from the hood due to the reverse flow was not obvious with the enclosure type in any condition, caution should be exercised to prevent exposure when the worker is seated. Regardless of the hood type, the increase in the capture velocity was effective in decreasing exposure due to the reverse flow.