We have shown previously that ascorbate has two distinct actions in the bovine perfused ciliary artery: a blockade of agonist (bradykinin or acetylcholine)-induced endothelium-derived hyperpolarising factor (EDHF) activity and an elevation of tone, resulting from blockade of flow-induced EDHF (Nelli et al., 2004, Stirrat et al., 2006). Both effects of ascorbate require the presence of flow and are abolished in the presence of zero or low flow. These studies were conducted using an open perfusion system where it was not possible to hold pressure constant when varying the rate of flow. The aim of this study was to make use of a pressure myograph to determine if the flow-dependent actions of ascorbate could still be elicited when pressure was held constant. Segments of bovine extraocular ciliary artery were cannulated at both ends, mounted in a pressure myograph and perfused at flow rates of 2.5, 5 or 10 ml min^-1 using a peristaltic pump. Pressure was held constant at 100 mm Hg by adjusting the height of the outflow tube and external diameter was monitored throughout. In some experiments EDHF activity was isolated by blocking nitric oxide synthase and cyclooxygenase using L-NAME (100 μM) and indomethacin (10 μM), respectively. Bradykinin (1 nmol) was injected into the perfusion fluid to evoke endothelium-dependent vasodilatation. In all experiments, n>6. When pressurised to 100 mm Hg the external diameter of ciliary artery segments was 1,609 ± 100 μm. Infusion of ascorbate (50 μM) induced constriction that was dependent upon the rate of flow: 1 ± 38, 113 ± 26 and 242 ± 52 μm at flow rates of 2.5, 5 and 10 ml min^-1, respectively. Treatment with L-NAME and indomethacin induced a constriction which did not vary with flow rate: 106 ± 24, 84 ± 30 and 131 ± 23 μm at flow rates of 2.5, 5 and 10 ml min^-1, respectively. Under these conditions, the ascorbate-induced constrictions were generally enhanced: 146 ± 55, 135 ± 33 and 457 ± 25 μm at flow rates of 2.5, 5 and 10 ml min^-1, respectively. Bradykinin-induced, EDHF-dependent dilatation did not vary with flow rate: 59.8 ± 4.7, 56.9 ± 4.1 and 51.8 ± 3.8 % at flow rates of 2.5, 5 and 10 ml min^-1, respectively. Ascorbate, reduced these dilatations in a flow-dependent manner to 41.4 ± 10.9, 39.2 ± 5.1 and 0 ± 0 % at flow rates of 2.5, 5 and 10 ml min^-1, respectively. These findings demonstrate that when the bovine ciliary artery is held at constant pressure, the ability of ascorbate to inhibit agonist- and flow-induced EDHF remains flow-dependent.