According to our data in rats, peripheral 3.4-dihydroxyphenylalanine (DOPA) kinetics are similar to striatal DOPA and dopamine kinetics. The measurement of plasma l-3.4-dihydroxyphenylalanine (L-dopa) concentration is thus useful to predict dopamine kinetics in the striatum and to treat the motor fluctuations of parkinsonian patients. In patients with Parkinson's disease (PD), long-term L-dopa therapy accelerated DOPA absorption and steepened features of L-dopa pharmacokinetics. In the senile-onset group, the pharmacokinetic pattern did not change even after long-term L-dopa therapy. The frequency of motor fluctuations is much lower in senile-onset patients with PD than in middle-onset patients. Differences in the pattern of L-dopa pharmacokinetics in the two groups may explain why the senile-onset group rarely develops ‘wearing-off', even after long-term L-dopa therapy. L-dopa is transported by a saturable active transporter system, called the LNAA (large neutral amino acid) system, in the gut and blood brain barrier. L-dopa absorption is thus affected by food intake, especially a proteinrich diet. The slope of the time-concentration curve for L-dopa administered before a meal is steeper than if it is administered after a meal. Considering that pulsative stimulation of L-dopa may cause motor fluctuations, L-dopa should be given after meals whenever possible, even if it necessitates a higher L-dopa dose.