In dogs anesthetized with pentobarbital, central vena caval pressure (CVP), portal venous pressure (PVP), and intrahepatic lobar venous pressure (proximal to the hepatic venous sphincters) were measured. The objective was to determine some characteristics of the intrahepatic vascular resistance sites (proximal and distal to the hepatic venous sphincters) including testing predictions made using a recent mathematical model of distensible hepatic venous resistance. The stimulus used was a brief rise in CVP produced by transient occlusion of the thoracic vena cava in control state and when vascular resistance was elevated by infusions of norepinephrine or histamine, or by nerve stimulation. The percent transmission of the downstream pressure rise to upstream sites past areas of vascular resistance was elevated. Even small increments in CVP are partially transmitted upstream. The data are incompatible with the vascular waterfall phenomenon which predicts that venous pressure increments are not transmitted upstream until a critical pressure is overcome and then further increments would be 100% transmitted. The hepatic sphincters show the following characteristics. First, small rises in CVP are transmitted less than large elevations; as the CVP rises, the sphincters passively distend and allow a greater percent transmission upstream, thus a large rise in CVP is more fully transmitted than a small rise in CVP. Second, the amount of pressure transmission upstream is determined by the vascular resistance across which the pressure is transmitted. As nerves, norepinephrine, or histamine cause the hepatic sphincters to contract, the percent transmission becomes less and the distensibility of the sphincters is reduced. Similar characteristics are shown for the "presinusoidal" vascular resistance and the hepatic venous sphincter resistance. Finally, a unit of pressure rise in downstream pressure will be more completely transmitted upstream as the basal starting downstream pressure is increased. These data fulfill the predictions of the distensible hepatic venous sphincter model developed for the cat liver and are incompatible with the Starling resistor – vascular waterfall theory. The distensible hepatic venous resistance allows the splanchnic blood volume to most efficiently buffer the largest changes in CVP by transmitting proportionately more pressure to the highly compliant splanchnic vessels. In addition the distensible sphincters serve to autoregulate portal venous pressure. As portal flow changes, the passively distensible sphincters minimize changes in PVP.