E Ef ff fe ec ct t o of f u up pp pe er r a ai ir rw wa ay y c co oo ol li in ng g a an nd d C CO O 2 2 o on n d di ia ap ph hr ra ag gm m a an nd d g ge en ni io o--h hy yo oi id d m mu us sc cl le e a ac ct ti iv vi it ty y i in n t th he e r ra at t Diaphragm and geniohyoid electromyographic activities were recorded in anaesthetized rats, breathing spontaneously through a low-cervical tracheostomy. Warmed, humidified air containing 0 or 10% CO 2 and cooled, room humidity air were applied at constant flow to the UA through a high-cervical tracheostomy. Spontaneous tracheal airflow, UA airflow and temperature, blood pressure, and rectal temperature were recorded.In all animals, the geniohyoid muscle had phasic inspiratory activity, which slightly preceded diaphragmatic activity. CO 2 had no effect on mean peak integrated diaphragmatic activity and variable effects on geniohyoid activity. The coefficients of variation of these activities were unaffected by CO 2 . Similar results were obtained following bilateral mid-cervical vagotomy. Cool air decreased respiratory frequency (78±8%) (mean±SD % of control), peak inspiratory flow (78±5%) and diaphragmatic activity (77±4%), and increased geniohyoid activity (149±11%). Cutting the superior laryngeal nerves abolished these effects.In conclusion, whilst moderate upper airway cooling inhibits breathing and excites geniohyoid muscle activity, upper airway carbon dioxide has minimal effect. Eur Respir J., 1996Respir J., , 9, 2323 Cooling of the upper airway (UA) stimulates laryngeal cold receptors [1] and inhibits laryngeal mechanoreceptors [2] running in the superior laryngeal nerve (SLN). These receptors are known to reflexly influence breathing and UA muscle activity, but the effects of cooling on UA muscle activity are controversial. In anaesthetized dogs, MATHEW et al. [3] found that laryngeal cooling (20-25°C) had no effect on posterior cricoarytenoid muscle activity but inhibited the muscle's response to UA occlusion, and this effect was abolished by SLN section. In anaesthetized cats, JAMMES et al. [4] reported that laryngeal cooling (8°C) inhibited laryngeal muscle activity (the identity of the muscle was not determined) but the effect of SLN section on this response was not examined. On the other hand, UKABAM et al. [5], in decerebrate, vagotomized cats, found that hypoglossal nerve activity was either excited or inhibited by laryngeal cooling (8-16°C) through a SLN-mediated reflex. The cause of these conflicting results may be due to whether the vagi were intact or cut, since vagal afferents greatly influence UA muscle activity [6], or else it may be related to the low and wide-ranging temperatures used. In anaesthetized rats, we have previously shown that more moderate cooling (25-30°C) of the UA causes a substantial fall in UA resistance, which is partly mediated through a SLN reflex [7]. The mechanism of this effect is unknown but we have speculated that it may be due to reflex activation of UA dilator muscles.The application of CO 2 to the UA in cats has ...