Twenty patients with Parkinson's disease were treated with the MAO-B inhibitor selegiline (l-deprenyl) and placebo without levodopa (L-dopa) in a randomized double-blind clinical cross-over study to analyze relative importances of dopamine (DA) synthesis and metabolism. The daily dose of selegiline was gradually increased to a maximum of 30 mg in all patients. The clinical neurological disability (Columbia score) was about 10% less on selegiline (30 mg/day) than on placebo. This difference was neither statistically nor clinically significant. The results are compatible with the possibility that treatment with selegiline without concomitant L-dopa does not significantly increase DA concentration which remains low and is determined mainly by tyrosine hydroxylase activity. At low DA levels the DA re-uptake mechanism recaptures most of the released DA and DA deamination is of minor significance. The pathway of DA oxidation becomes more important only at higher DA concentrations, accomplished by bypassing the rate limiting step of tyrosine hydroxylation using L-dopa.