The efficiency of an industrial local exhaust hood was evaluated by a tracer gas technique. Sulfur-hexafluoride (2551624) (SF) was sampled in the duct, and its concentration was determined by gas chromatography (GC). SF was continuously released under controlled conditions from an SF source located in the hood duct inlet. The SF source was then positioned at various locations around the hood or at locations where industrial contaminants might originate. SF was again measured in the duct. The concentration found was compared with that released in the hood duct inlet. The ratio of these values (expressed in percent) was considered as the hood efficiency at a specific point. Three basic parameters were critical to the proper evaluation of the technique: proper SF detection conditions, proper sampling location, and proper SF discharge conditions. An increase of the GC column temperature, up to 100 degrees-C, resulted in a shorter elution time for oxygen peak; SF elution time remained virtually unchanged. Changes of pressure across a glass fiber filter connected to the GC column did not significantly affect SF peak height. The duct center line was determined as the best position for the sampling probe. A lower SF concentration was found in the duct where the tracer discharge velocity was higher than that of the air flow. The author concludes that the tracer gas technique is capable of directly evaluating the hood efficiency associated with fumes, gases, and vapors or very fine particles.