The effect of facial attachments on breathing was measured by respiratory induction plethysmography (RIP) during quiet sleep in 32 studies in 18 infants. The addition of a face mask plus pneumotachograph led to an increase in tidal volume (VT) (22.0 +/- 13.5%, p less than 0.01) during 5-min sleep studies when compared to measurements using RIP alone. Applying only the mask rim also led to an increase in VT (14.6 +/- 3.1%, p less than 0.05). A significant increase in VT was noted in 3 of 6 infants studied when a lightweight cardboard ring was in place perinasally. Respiratory frequency fell significantly in the mask/pneumotachograph group (-5.9 +/- 10.0%, p less than 0.05) and with the mask rim (-7.4 +/- 8.8%, p less than 0.01), but there was individual infant variation. Minute ventilation rose significantly (19.1 +/- 16.9%, p less than 0.01) only with the addition of the mask and pneumotachograph. Instrument deadspace can account for some of the increase in VT noted, but in its absence, sensory stimulation of the trigeminal area can augment tidal breathing.