Pharyngeal dilator muscles are important in the pathophysiology of obstructive sleep apnoea syndrome (OSA). We have previously shown that during wakefulness, the activity of both the genioglossus (GGEMG) and tensor palatini (TPEMG) is greater in patients with OSA compared with controls. Further, EMG activity decreases at sleep onset, and the decrement is greater in apnoea patients than in healthy controls. In addition, it is known that the prevalence of OSA is greater in middle-aged compared with younger men. Thus, we had two goals in this study. First we compared upper airway muscle activity between young and middle-aged healthy men compared with men with OSA. We also explored the mechanisms responsible for the decrement in muscle activity at sleep onset in these groups. We investigated muscle activity, ventilation , and upper airway resistance (UAR) during wakefulness and sleep onset (transition from alpha to EEG activity) in all three groups. Measurements were obtained during basal breathing (BB) and nasal continuous positive airway pressure (CPAP) was applied to reduce negative pressure-mediated muscle activation). We found that during wakefulness there was a gradation of GGEMG and UAR (younger < older < OSA) and that muscle activity was reduced by the application of nasal CPAP (to a greater degree in the OSA patients). Although CPAP eliminated differences in UAR during wakefulness and sleep, GGEMG remained greater in the OSA patients. During sleep onset, a greater initial fall in GGEMG was seen in the OSA patients followed by subsequent muscle recruitment in the third to fifth breaths following the alpha to transition. On the CPAP night, and GGEMG still fell further in the OSA patients compared with control subjects. CPAP prevented the rise in UAR at sleep onset along with the associated recruitment in GGEMG. Differences in TPEMG among the groups were not significant. These data suggest that the middle-aged men had upper airway function midway between that of young normal men and the abnormal airway of those with OSA. Furthermore it suggests that the initial sleep onset reduction in upper airway muscle activity is due to loss of a 'wakefulness' stimulus, rather than to loss of responsiveness to negative pressure, and that this wakefulness stimulus may be greater in the OSA patient than in healthy controls.