A new approach to clinical olfactometry is presented using nasal sprays. A special dosage valve was used to standardize an aerosol size to 40 microns. For evaluation a 6 x 6 matrix (substance/olfactory quality) with verbal associative clues was used according to test psychological findings. Validation took place in three steps after determining smell-associated thresholds in preliminary examinations. Recognition of different aromas administered either by spraying into the open mouth from a distance of 5 cm using nasal sprays or sniffing opened bottles was compared in 50 subjectively normosmic subjects. Findings showed that the correspondence between the two procedures was relatively high, with global identification of odorous substances as high as 98.4% in the spray test. The number of substances used was next reduced to those six substances providing the most reproducible results in an investigation with 56 normosmic and 55 hyposmic subjects. Verbal associative clues were also tested. In a last step 50 patients with hyposmia of various origin and 110 normosmic subjects were tested, allowing previous results to be reproduced regardless of the cause of hyposmia. The spray test was shown to be easily performed and was suitable as a screening test, with a sensitivity of 88% and a specificity of 100%.