1. The effect of chlorpromazine (CPZ) on the metabolism of dopamine and 5-hydroxytryptamine in dog brain was investigated by following the concentrations of the acid metabolites of these amines, homovanillic acid, 3,4-dihydroxyphenylacetic acid and 5-hydroxyindolylacetic acid, in the ventricular cerebrospinal fluid (C.S.F.) of dogs over a period of 5 hr after intravenous administration of CPZ (2.5, 5, 10 and 15 mg/kg), using the technique of serial sampling of lateral ventricular C.S.F. "Low" doses (2.5-10 mg/kg) produced a rise in the concentration of homovanillic acid and smaller increases in the concentrations of 3,4-dihydroxyphenylacetic acid and 5-hydroxyindolylacetic acid. "High" doses (10-15 mg/kg) had a lesser effect on the concentration of homovanillic acid and had no effect on, or decreased, the concentrations of 3,4-dihydroxyphenylacetic acid and 5-hydroxyindolylacetic acid. The concentration of 3,4-dihydroxyphenylacetic acid was maximal in the ventricular C.S.F. 2 hr after CPZ 5 mg/kg and was unaltered from the control level 2 hr after 15 mg/kg.2. The effects on the metabolism of brain amines of CPZ (5 mg/kg), doses which the serial sampling of C.S.F. experiments had indicated as producing maximal and minimal effects on dopamine metabolism in brain tissue, were studied by estimating the concentrations of adrenaline, noradrenaline, dopamine, metanephrine, methoxydopamine, homovanillic acid, 3,4-dihydroxyphenylacetic acid and 5-hydroxyindolylacetic acid in the hypothalamus, midbrain, thalamus, hindbrain, cortex, globus pallidus and caudate nucleus of control dogs and of dogs treated with CPZ intravenously 2 hr before killing. The concentrations of homovanillic acid, 3,4-dihydroxyphenylacetic acid and 5-hydroxyindolylacetic acid were estimated in samples of ventricular C.S.F. withdrawn from these dogs 2 hr after the injection of CPZ (i.e., immediately before death).3. The following changes in concentrations were observed. Dopamine: CPZ 5 mg/kg produced no change in the concentration in the caudate nucleus, globus pallidus and midbrain and increased the concentration in the thalamus; CPZ 15 mg/kg appeared to cause a reduction in the concentration of this amine in the caudate nucleus and globus pallidus. Homovanillic acid and 3,4-dihydroxyphenylacetic acid: CPZ 5 mg/kg increased the concentrations of both acids in the caudate nucleus and had no effect on the concentrations of the acids in the globus pallidus, hypothalamus and thalamus; CPZ 15 mg/kg produced no change in the concentrations of the acids in any area of the brain. Methoxydopamine: CPZ 5 mg/kg and 15 mg/kg reduced the concentration in the caudate nucleus. Noradrenaline: The concentrations in the hypothalamus, midbrain, thalamus and hindbrain were slightly increased by CPZ 5 mg/kg and 15 mg/kg. Only in the thalamus was a statistically significant increase in noradrenaline observed.4. It was concluded that the actions of chlorpromazine on catecholamine synthesis and metabolism in the brain of the dog are dose dependent. A dose of CPZ 5 mg/kg was postulated to have the following actions: (i) to increase dopamine synthesis; (ii) to activate mitochondrial monoamine oxidase. A dose of CPZ 15 mg/kg was postulated to act as follows: (i) to decrease dopamine synthesis; or (ii) to release dopamine from its storage sites.5. The ratios of the concentrations of homovanillic acid, 3,4-dihydroxyphenylacetic acid and 5-hydroxyindolylacetic acid in the caudate nucleus to the concentrations of these acids in the ventricular C.S.F. were the same in the control dogs as in the dogs treated with CPZ (5 mg/kg and 15 mg/kg). It was concluded that the levels of the acid metabolites of dopamine in lateral ventricular C.S.F. reflect the levels of these acids in the caudate nucleus.