The ability to maintain pharyngeal patency is compromised in infants who have apneic episodes associated with airway obstruction. Since the genioglossus (GG) muscle is thought to be important in maintaining pharyngeal patency, we measure the GG EMG with sublingual surfact electrodes during unobstructed breathing and in response to end-expiratory airway occlusion. Studies were performed in nine premature infants with mixed and obstructive apnea and in eight nonapneic control infants. Phasic GG EMG was usually absent during normal tidal breathing in both groups of infants, however, GG activity typically appeared during airway occlusion. The response of the GG muscle during airway occlusion differed between control and apneic infants. During the first three occluded inspiratory efforts, control infants had 42 ± 5, 74 ± 5, and 80 ± 5% (mean ± SEM) of their occlusions associated with a GG EMG response, respectively. In contrast, apneic infants had significantly fewer (13 ± 4, 38 ± 9, and 52 ± 9%) occlusions associated with a GG EMG response. There was a delay in onset of the GG EMG when compared to the onset of the diaphragh EMG and initial negative esophageal pressure swing, but this delay decreased with each subsequent appearance of the GG EMG in both infant groups. Infants with mixed and obstructive apnea thus have decreased activation of their GG in response to occlusions which may reflect their inability to recruit dilating muscles of the upper airway during spontaneous airway obstruction.