The effects of dopamine blockade on the endocrine and metabolic response to starvation have been investigated by administration of metoclopramide, 30 mg daily, or placebo to five normal subjects fasted for sixty hours on two occasions. Blood glucose and alanine concentrations fell with starvation and metoclopramide had no further effect. Concentrations of the other gluconeogenic precursors, lactate and pyruvate, were also unaffected by metoclopramide. The rise in circulating ketone body concentrations with fasting was impaired by metoclopramide, significantly from 44 h onwards (blood total ketone body concentration at 60 h, 3.42 +/- 0.94 mmol/l with placebo; 2.08 +/- 0.67 mmol/l with metoclopramide, P less than 0.05). Blood glycerol and plasma non-esterified fatty acids (NEFA) levels rose with starvation, and metoclopramide had no further effect. Serum insulin concentrations remained low with fasting, while circulating glucagon and growth hormone levels rose. Similar changes were noted with both metoclopramide and placebo. Serum prolactin concentrations during starvation were elevated two to four fold by metoclopramide. The inhibitory effect of dopamine blockade on ketosis thus occurred despite hyperprolactinaemia, and did not result from measurable alterations in insulin, glucagon or growth hormone secretion. The data suggest a stimulatory role for endogenous dopamine on starvation ketonaemia in man.