An extract from a dry young wine from Maccabeo was studied by aroma extract dilution analysis (AEDA), quantitative gas chromatography, and different sensory studies. In a first study, 53 different aroma compounds were quantified and used to prepare aroma models. 2-Methyl-3-furanthiol (FD = 16) and 4-methyl-4-mercaptopentan-2-one (FD = 2), could not be quantified and were not included in those models, which were not very similar to the original wine. Omission tests did not show the existence of impact compounds. In another set of experiments, selected aroma chemicals were added to the original wine, but in only in two cases (isoamyl acetate and gamma-nonalactone) was a positive effect noted, on banana and citric notes, respectively. After these discouraging results, 4-methyl-4-mercaptopentan-2-one and 2-methyl-3-furanthiol were quantified and included in the models. The concentration of the former was as low as 5 ng x L(-)(1) (odor threshold = 0.8 ng x L(-)(1)); however, its inclusion in the synthetic mixture had a significant effect, making it very close to the original wine. Its role was confirmed by omission tests. Results are briefly discussed.