It has been proposed that exercise induced asthma is a result of "rapid expansion of the blood volume of peribronchial plexi" (McFadden ER, Lancet 1990;335:880-3). This hypothesis proposes that the development of exercise induced asthma depends on the thermal gradient in the airways at the end of hyperpnoea. The events that result in exercise induced asthma are vasoconstriction and airway cooling followed by reactive hyperaemia. We agree that the airway microcirculation has the potential for contributing to the pathophysiology of exercise induced asthma. We do, however, question whether reactive hyperaemia, in response to airway cooling, is the mechanism whereby hyperpnoea provokes airways obstruction in asthmatic patients. Further, we question whether vasoconstriction accompanies dry air breathing and whether an abnormal temperature gradient and rapid rewarming of the airways are prerequisites for exercise induced asthma. From published experiments we conclude that dry air breathing is associated with vasodilation and increase in airway blood flow rather than vasoconstriction and a decrease in blood flow to the airways. We propose that the stimulus for the increase in airway blood flow is an increase in osmolarity of the airway submucosa. This osmotic change is caused by the movement of water to the airway lumen in response to evaporative water loss during hyperpnoea. The increase in airway blood flow may occur directly or indirectly by the osmotic release of mediators. Exercise induced asthma is most likely to be due to the contraction of bronchial smooth muscle caused by the same mediators. Whether it is enhanced or inhibited by alterations in airway blood flow is not yet established in man.