The parallel Lack system is a new modification of the Mapleson A system comprising separate inspiratory and expiratory tubes. To determine that the function of the system was that anticipated of a Mapleson A, the fresh gas flow requirements to prevent rebreathing during spontaneous ventilation were assessed in three situations: (1) a lung model (2) conscious volunteers and (3) anaesthetised patients. Two sets of criteria to define rebreathing were used; (A) those based on changes in ventilation or end-expired carbon dioxide tension and (B) minimum inspired carbon dioxide tension. Using A, rebreathing occurred at a fresh gas flow to minute ventilation ratio (VF/VE) of 0.75 for the lung model, and 0.73 for conscious volunteers. These results were comparable to those obtained for a Magill attachment. They were also close to the point at which mechanical dead space began to increase in the lung model. Criteria B gave much lower values for the onset of rebreathing. Rebreathing was present by criteria A in five of the six anaesthetised patients at a fresh gas flow of 60 ml.kg-1.min-1 (VF/VF of 0.78). The results confirm that the parallel Lack behaves as a Mapleson A system. The resistance to breathing posed by the parallel Lack was also comparable to the Magill system.