Experiments were performed on two groups of 42-h-fasted conscious dogs (n = 6/group). Somatostatin was given peripherally with insulin (4-fold basal) and glucagon (basal) intraportally. In the first experimental period, glucose was infused peripherally to double the hepatic glucose load (HGL) in both groups. In the second experimental period, glucose (21.8 micromol. kg-1. min-1) was infused intraportally and the peripheral glucose infusion rate (PeGIR) was reduced to maintain the precreating HGL in the portal signal (PO) group, whereas saline was given intraportally in the control (CON) group and PeGIR was not changed. In the third period, the portal glucose infusion was stopped in the PO group and PeGIR was increased to sustain HGL. PeGIR was continued in the CON group. The glucose loads to the liver did not differ in the CON and PO groups. Net hepatic glucose uptake was 9.6 +/- 2.5, 11.6 +/- 2.6, and 15.5 +/- 3.2 vs. 10.8 +/- 1.8, 23.7 +/- 3.0, and 15.5 +/- 1.1 micromol. kg-1. min-1, and nonhepatic glucose uptake (non-HGU) was 29.8 +/- 1.1, 40.1 +/- 4.5, and 49.5 +/- 4.0 vs. 26.6 +/- 4.3, 23.2 +/- 4.0, and 40.4 +/- 3.1 micromol. kg-1. min-1 in the CON and PO groups during the three periods, respectively. Cessation of the portal signal shifted NHGU and non-HGU to rates similar to those evident in the CON group within 10 min. These results indicate that even under hyperinsulinemic conditions the effects of the portal signal on hepatic and peripheral glucose uptake are rapidly reversible.