This paper presents results of an experimental study into factors contributing to decay of odor samples during storage, between 4 and 40 hr after sample collection. The odor studied was sampled from a tobacco processing plant as part of collaborative research with a view to establishing a manual outlining methods for odor annoyance management, specifically for the tobacco industry. In August and September 1997, an experimental program was carried out in which two types of tobacco odor were sampled: Burley Toaster and Mix. The dependent variable was odor concentration in the bag, measured by dynamic olfactometry in accordance with the draft Comité Européen de Normalisation (CEN) standard EN13725 “Air Quality— Determination of Odor Concentration by Dynamic Olfactometry.” The independent variables were sampling bag material, degree of dilution during sampling, dilution gas used, particle removal during sampling, and age of sample in hours. In the first phase, 94 odor analyses were carried out. In a second test, 32 samples were analyzed for odor concentration. In addition, 16 samples were analyzed using gas chromatography-mass spectrometry (GC-MS). Analysis of the results (analysis of variance) led to the unexpected conclusion that Nalophan film bags performed significantly better than metalized Cali-Bond layered film as a bag material. The odor concentration of samples in Nalophan bags remained relatively stable between 4 and 12 hr after sampling. After 30 hr, decay to about half the initial concentration, as measured at 4 hr, was observed. Particle removal during sampling caused the odor concentration in the bags to be reduced by approximately 20%. For practical reasons, particle removal remains useful, to avoid contamination of equipment. Using air or nitrogen as the neutral gas for pre-dilution during sampling or the dilution factor used (between factor 2 and 6) did not appear to have an effect on the decay characteristic of odor samples. The following recommendations are suggested for the practice of collecting odor samples and apply specifically to tobacco processing emissions: • Analyze samples as soon as possible, preferably within 12 hr; • When samples age for more than 12 hr, decay is likely to cause a reduction in odor concentration to half the original concentration at age 30 hr; • Use sampling bags made of Nalophan NA or benchmark performance of other materials against Nalophan NA before using alternative materials; • Use pre-dilution when sampling only for the purpose of avoiding condensation during sample storage. Use an appropriate minimum dilution factor to avoid condensation; • Both nitrogen and high-purity (synthetic) air are suitable to use as neutral gas for pre-dilution; and • When sampling tobacco odors, use an odorless filter to remove particles. This practice removes a source of variation and avoids contamination of equipment. The effect on results, despite being consistently lower in odor concentration, is not meaningful in terms of perceived intensity or annoyance potential.