The effects on breathing of (1) nose-clips, with and without mouthpieces, (2) tube breathing with incremental dead spaces up to 600 ml, and (3) a Venturi mask with a high air inflow reducing the effective dead space to zero were measured with respiratory inductance plethysmography. The effective dead spaces of the lengths of tubing were 25% smaller than their water-displacement volumes. The principal response to the increase in dead space was an increase in VT without any change in frequency or timing. The increase in VE was 16% greater than the ventilation of the effective dead space. The facemask with a high flow of air did not affect the pattern of breathing. External devices alone (nose-clips, mouthpieces) by themselves brought about a significant increase in VT (+10%, P less than 0.05) and a small drop in frequency, which in the case of a nose-clip alone was large enough to be significant (-15%, P less than 0.05). The increase in DS with the change from nasal to oral breathing may have contributed to the accompanying increase in VT; the fall in frequency with the nose-clip alone might have been caused by stimulation of receptors in the trigeminal area.