Introduction The literature on the subject of the effect on blood flow of vessel stenosis reveals results and conclusions ranging widely from Shipley and Gregg's5 statement that they found no justification for the contention that a rather marked degree of external constriction is required to produce a significant reduction in flow through a vessel, to the demonstration by Cannon et al1 that a 99.1% reduction in lumen area produced a mere 30% fall in flow. May, DeWeese, and Rob4 have summarized work in this field and find that with a 1-cm-long stenosis the luminal area in the canine iliac artery required reduction to 20% of normal to begin to decrease blood flow. The variation in results previously reported may be due to many factors, such as failure to monitor viscosity changes in the dog's blood, inadequate control or record of pressures (Mann et al3), unreliable flow