The paper by M. A. Perry et al (1) constitutes an important contribution for several reasons. Besides adding to basic knowledge of intrinsic blood flow mechanismS, this paper is relevant to clinical research. Injury in the gastric mucosa can be caused by an increase in the ratio of acid back-diffusion/ mucosal blood flow (2). Patients at risk for stress gastritis are frequently given wholesale amounts of cimetidine to reduce gastric acidity. But, if gastric mucosal autoregulation is a function of metabolic activity, then inhibition with cimetidine may deprive the mucosa of a critical defense mechanism against ischemic injury. The clinical relevance of the data of Perry et al must be tempered by the "uncertainty principle" of physiology, ie, "the more precisely we measure it, the more likely we distort it." Besides recruitment of unperfused tissue, which was discussed, other possible sources of artifact are worthy of discussion: (a) Anesthesia, surgical trauma, and fluid loss (3) are factors which may change circulating catecholamines and other agents capable of influencing local autoregulation. (b) Arterial perfusion Via tubing is known to change vascular resistance (4). Injured platelets may release several vasoactive compounds (serotonin and prostaglandins). These agents may profoundly affect local autoregulatory responses. (c) Heparin infusion is also known to cause platelet aggregation and subsequent release of these vasoactive compounds (5). This may deplete the platelets and prevent them from playing a role in local autoregulation. Perhaps platelets react to some local stimulus and release prostaglandins to mediate local vasodilation, or perhaps, platelet release of serotonin is important for enhancing the effects of catecholamines in local (a few dozen capillaries) vasoconstriction. (d) Recent studies on histamine have shown that H1 receptors may have a direct vasoconstrictor effect in the gastric vasculature (6). This effect is overshadowed by the vasodilator and stimulatory effects of the H2 receptors.