Visualization of the head of the pancreas by CT was prospectively evaluated in two groups of 100 patients who did not have pancreatic disease. Patients were given either a fat-density oral contrast material (12.5% corn-oil emulsion and metoclopramide) or a conventional high-density oral contrast material (barium suspension or iodinated solution). There was no statistically significant difference in the subjects' tolerance to the two regimens. There was, however, a significant improvement in ability to distinguish the head of the pancreas from the duodenal C-loop when the fat-density contrast material was given. When pancreaticoduodenal discrimination was graded, patients given corn-oil emulsion and metoclopramide received an average score of 0.94, whereas those given the high-density agent received an average score of 0.74, with 1.00 being the highest possible score (p less than .005). These data suggest that for routine CT evaluation of the head of the pancreas, a combination of corn-oil emulsion and metoclopramide may be superior to the conventional high-density oral contrast agents given without metoclopramide.