108 Brain Research, 85 (t975) 108-113 ;© Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands Olfactory stimulation with controlled and monitored step pulses of odor JOHN S. KAUER ANt) GORDON M. SHEPHERD Department of Physiology, Yale University School of Medicine, New Haven, Conn. 06510 (U.S.A.) (Accepted November 1st, 1974) Analysis of the neural basis of olfaction has been impeded by lack of ability to deliver discrete stimuli such as in the analysis of sensory mechanisms in vision 5, audition 11, and somesthesis 6. However, recently it has been possible to achieve punctate odor stimulation of the olfactory receptor surface and to obtain evidence for receptive field organization in the olfactory pathway 4. As a further advance in the control of odorous stimuli, we have developed a method for delivering odors as step pulses which are monitored at the site of stimulation. The experimental set-up consists of 3 main components: (1) an olfactometer which controls the flow rate and concentration of the stimulus; (2) a delivery system which controls the onset, duration and termination of the stimulus, and (3) a moni- toring system which measures the time course of the pulse containing the odor. The olfactometer has been previously described by Kauer 3. As adapted for the present experiments its output consisted of an odor of known concentration in a carrier gas composed of 95 ~ 02 and 5 ~o CO2. This odorized stream was led to the middle of 3 concentric glass pipettes which made up the delivery system (see Fig. 1A). Between stimulus applications, the odor was withdrawn by a vacuum in the central pipette (solid arrows in Fig. 1A) and exhausted outside the room and thus did not reach the preparation. Delivery of the stimulus to the olfactory mucosa was accomplished by activating a solenoid valve which switched the vacuum from the center pipette to the outermost pipette. The solenoid was triggered by a conventional electrophysiological stimulator which controlled the duration of the pulse. In this way the odor was rapidly switched onto the preparation, while the vacuum in the outer pipette prevented lateral odor spread (see broken lines in Fig. IA). Thus an abrupt onset and termination of the stimulus was achieved by a switching system that was remote from the delivery pi- pettes. For monitoring the step pulses of odor, several systems have been tested; these included the flame ionization detector of a gas chromatograph 3 (see also ref. 9), a mass spectrometer, a thermistor, and a mechanical flow detector. Judged by criteria of rapid response time, stability, accuracy, and simplicity, each of these methods was unsuitable in one or more respects. The system that has been most satisfactory